Decoding Words

  The place of phonics in teaching reading has always been a hot issue. And, as with most volatile topics, intelligent answers aren’t quite as simple as the questions. I hope to help you make sense out of what you hear, and to give you a basis for sound planning in teaching your own child.
  Phonics is the relationship between the individual sounds in a word and the letters that represent them. Everyone agrees that readers need to understand how sounds relate to words. The issue involves how this should be taught. To explore this area, let’s first trace the history of education as it involves reading. (You could skip it if you are in a hurry). Then we will look at the options for teaching the basics of how to read.

How the curriculum has changed

  Looking back helps us view modern pressures on the curriculum in perspective. An expert in the history of education might see events a little differently from the way I sketch them below, but this can give you a fair idea.
  l  Education for the European elite class during the Middle Ages is described in my chapter, “Structure for Learning.”
  l  Not until after the US civil war in the 1860s did education slowly  begin to become widely available to common people. I explain a little of this history in the chapter, “Keeping Peace With School Authorities.”
  l  At the entrance of the 20th century, elocution — or public speaking employing gestures and vocal delivery — was considered an important art. Thus the reading program in schools employed oral reading with inflection, distinct pronunciation and, of course, phonics. This was McGuffey’s time.
  l  A change occurred when good methods of standardized testing were developed, and educational measurement promised to find scientifically accurate answers to every problem. Researchers discovered that students well-trained in oral reading tended to read slowly. Silent reading then became a primary concern, and teaching techniques began emphasizing quick word recognition. The new challenge was “see and say,” rather than “sound it out.”
  l  Then the project method came into vogue. Instead of the various school subjects taught one at a time in isolation, projects were chosen that required application of math, reading, social studies, science, and so on. Learning was shown to have a purpose beyond pleasing the teacher and getting good marks on a report card. Research studies showed better achievement from schools trying the new method, likely influenced by the higher expectation of the students involved.
  Teaching through projects (reincarnated recently as the unit study method) is still a good idea and works well for home schools. That is, it’s good if you are alert and avoid its pitfalls. For one thing, projects don’t tend to foster the reinforcement of basic skills so essential to thorough learning and remembering. Other points are covered elsewhere in this book.
  l  As the project method was adopted far and wide, basic skills, including reading, began to suffer. Neither word recognition nor phonics drill was getting much attention.
  l  In the 1930s and again after World War II, schools were loudly criticized (although students were learning well) and they turned back to more basic skill-building.
  l  In 1955 Rudolph Flesch published his book, Why Johnny Can’t Read. The answer? Neglect of phonics instruction. A sequel appeared in 1979.1
  l  Then in the 60s the cry became, “We want it relevant,” and drill (and, to some degree, phonics) wasn’t.
  l  In the 70s national test scores began to slip and the watchword was “back to the basics.”
  l  The 80s could be characterized by a call for life application and realistic preparation for a changing job market. Skills for processing numbers and words were emphasized a little less.
  l  In the 90s, the emphasis has been on usefulness of learning as indicated by abilities achieved. The call for the various states to specify outcomes in their lists of achievement objectives has been viewed with suspicion by many who are concerned about public education. I discuss reasons for caution in the chapter, “Developing an Educational Framework.”
  l  In reading, “whole language” has been popular in many schools but has gotten a bad reputation and phonics is again in vogue.

  I have oversimplified some of this so you can see the bigger picture. The educational pendulum swings back and forth. Good school teachers have always worked for what they feel is best for kids and home teachers should, too.
  Incidentally, I don’t believe that a lack of relevance or a lack of basics or a lack of phonics or too much phonics or prescribed outcomes has been the real problem in schools. Drugs, sex, alcohol, tobacco, violence, TV, broken homes, teacher unrest, and similar influences are certainly more prevalent now than in the past and can’t help but interfere with the operation of schools.
  In spite of this, national test scores have been steadily improving, perhaps due to better methodology. I discuss this in chapter 2.

Clear thinking about phonics arguments

  What is the great point of controversy about how children should be introduced to reading? Some feel that decoding skills, specifically phonics, should be taught thoroughly as a foundation before any reading occurs beyond words or sentences for which the rules have been learned. By blending the succession of sounds made by letters or letter groups children are to be able to decode words and become good readers. Without this, it is felt that they will be poor readers and spellers.
  At the other end of the argument are those who feel children should, with a few words recognized from sight, begin reading words as they naturally appear in stories. This is believed to help them learn better and faster without sacrificing the knowledge of the combinations of letters that make up the words. New words are recognized by comparing them with known words.
  I began a serious investigation of this whole topic in the early 80s soon after I began  writing about home education. Although I make a conscious effort to be objective, I didn’t begin my study of the reading-methods debate without some ideas. I felt children should be taught a modest amount of phonics as the key to decoding words. My wife is a better reader than I am, and she studied phonics as a child. I vaguely remember a little phonics instruction and more clearly remember learning whole words. For both of us, this was in the early 40s. This is good personal-experience evidence, although with a rather small sample size for research reliability. Incidentally, my reading skill has gradually improved. It took a jump during the time of my doctoral studies in the late 60s and again in the 80s.
  As mentioned earlier, kids are amazingly adaptable. If we can motivate them, they will learn even with faulty methods. Still, there are principles for good learning and we all want the best for our kids. This a call to listen with both ears, separating fact from inference and trends from anecdotes.
  As we consider a topic of this nature, we need to cut through the emotional level of favorite ideas and sensationalism. Reading methods are sometimes denigrated or elevated on the basis of claims about sinister motives of teacher educators and about particular methods of instruction. Thus people are often frightened into choosing or avoiding particular materials or methods.
  Let’s look at the options beginning with specific phonics approaches and ending with a plan developed in recent years.

Some home schoolers, concerned about giving their children a solid reading foundation, like to use reprinted editions of the old McGuffey readers, evidently feeling that in the “good old days” educators really knew how to teach. McGuffey used a phonics approach although he considered whole word memorization to be an option.2
For several editions of The Home School Manual prior to this one, I have included a specific phonics methods chapter by Frank Rogers — with no regrets. Let’s consider the case he makes for what he calls "vertical phonics." Rogers’ system follows some of the principles in the Writing Road to Reading method. He believes, differently, however, in many ways including, # that reading is the road to writing, # that children should learn the names of the letters before phonics instruction; and # that word decoding can start with knowing the sounds of just four letters.
  He is very methodical. For him, the child is to be taught all the significant sounds of each vowel, each consonant, and certain letter combinations in the order of how frequently they occur in general reading material. For example, to discover the correct sound of the letter a in a particular word, the first sound tried is the one a makes in the word, “at.” If that sound combined with the other letters doesn’t make a word he or she recognizes, the sound of a in “ate” is tried; then its sound as in “all.” The a sounds in “again” and “any” must be treated as exceptions.
  In the more popular phonics method, the short vowel sounds (which occur in more words than others) are taught first. To Rogers, this is like learning the names of the presidents by memorizing all their first names, then going back to learn the last names. We have had some good discussions about how learning presidents is different from learning words, so he may hesitate to use that analogy. But learning presidents’ names together does describe his system. As he sees it, vertical phonics is first logical, then it works so it must be the best.
If we change uniforms and join the ranks of believers in the more popular phonics methodology, we find a defensible logic fortress there, too. By first teaching the short vowel sounds — the ones occurring most often — kids can decode more words earlier. The logic of this plan isn’t as nice, but the plan seems more natural, with simpler attack rules. Without solid research, we cannot say which is better or if it matters. As far as I know, kids learn well either way.
Practicing with word families is another approach. Certain letter combinations are learned by lists like “Sam, ham, jam, dam slam.” This method, too, is backed by good learning rationale, especially if taught with rhyming stories. In one of Bob’s Books booklets, which I have reviewed, I remember two lines in a story: “Into the tent the ten men went. Then to bed the ten were sent.” Even without cute stories, the supposedly dull lists of words hold a pleasant challenge for the young student. In reading “mop, drop, stop, crop, pop” I can see how the easy decoding allows the child to focus on word differences while the “op” word-pair sound settles into the subconscious.
Advocates following these three approaches feel, to varying degrees, that words should not be taught before the rules that govern them. This has a certain logic, but so does the argument that Mark Thogmartin uses. As you can see from his chapter, He uses children’s story books. The teacher sits with the child and, when a word appears that doesn’t make sense, assistance is provided to help analyze it, thus teaching a sort of phonics. Interest in the story provides incentive and a context for remembering the new word as well as the letter patterns in it.
    Neither analogies nor success stories are reliable evidence for a “best” way to learn to read. They may appear convincing but we need to be careful in accepting them as arguments.
 

Phonics and new information

  From reading Rudolph Flesch’s 1979 book, Why Johnny Still Can’t Read, I learned that the phonics issue was emotional and political. The educators (and politicians) hadn’t listened much when his earlier book came out in 1955 and again he was taking his case to the people. Unfortunately, politics still drives reading methodology in US schools. More on this later.
  I have questioned the relative importance of one reading approach over another as discussed earlier in this chapter. Now, just as I’m already late in getting this seventh edition to press, I discovered some information you may not have known about. I’d like to share some of it.
  Phonics offers a nice way to analyze words and, as we have seen. It follows logic. But some have questioned whether it provides a suitable model for effective teaching.
  In the 60s the reasonableness of phonics instruction was questioned by Theodore Clymer.3 He took four popular reading programs and identified 45 clearly stated phonics generalizations in them. Then he looked at the words used in those programs. Among the rules he listed as being taught in the four programs is “When there are two vowels, one of which is a final e, the first vowel is long and the e is silent.” It was seen to work with bone, came, date, and time. but not with done, come, have, and were. This is not too surprising because there are exceptions to phonics rules (all but one of his 45 have exceptions). The problem Clymer saw was that more than a third of the words in the texts classified as exceptions to this rather fundamental phonics rule just described! In fact, of the 30 rules involving vowels, only half worked 60% or more of the time in the reading texts. The more reliable phonics rules tend to apply to the less frequently appearing patterns. Several other researchers did studies to test his results and got similar results.4
  At about the same time, other researchers were asking if children were even able to identify the phonemes. D.J. Bruce5 found that none of the five- and six-year-old children in his study were able to remove a phoneme from a given word to make a new word. For example, they could not take the word fork, remove the /k/ sound, and say the new word for. Seven-year-olds succeeded on an average of 2 out of the 30 words. At eight and nine years (by the time they would have been reading well) the kids could score 50%.
  How can children figure out new words in a story without sounding them out? In 1965, Kenneth Goodman6 took a group of children to see if they could recognize words in context. For each child, he found a list of words on which only a few words were not recognized. Each child was then given a story which used all the words on his or her list. Instead of the unrecognized words, only blank lines appeared. When reading the stories they were able to recognize, from the context, most of the words they had not known in isolation. If they had used letter-sound correspondences to read the words, they should have done at least as well reading the words on the list. One might think children would learn errors this way or that they would not easily remember the new words, but this does not appear to be the case. Children are guided in analyzing letter sounds for these words as they are in phonics methods.
  An interesting comparison between whole language (learning from context as well as other cues) and phonics instruction appeared in the statistics from The National Assessment of Educational Progress (NAEP) test of fourth-grade readers.7 Teachers were asked to identify their teaching method as "whole language" or "phonics" and the degree of emphasis they gave it. There was no apparent difference between student scores of "moderate" emphasis teachers. However, scores of students with "heavy" emphasis teachers of whole language were 9% higher than those of heavy emphasis phonics teachers. Other studies have shown similar results but are not as impressive statistically because they were conducted with specific classes rather than with the masses of kids across the nation in the NAEP. We must also not miss the point here that, although not scoring quite as well, children with phonics instruction were learning reasonably well.
  See the notes at the end of this chapter for more about phonics and whole language.8 It is not a matter of one all good and the other all bad.
  How are children guided in learning from continuous (whole language) text? v We described the key word method in the reading chapter. v "Shared reading" is another technique. A story where illustrations that show the story are discussed, then the teacher reads it pointing to the words. After several readings, over a period of days, children memorize the language and "read" to the teacher and to each other as they figure out the printed words. I have seen a variation of this using Bible passages with words to be learned shown in bold face. After reading a passage several times, the teacher can stop at the special words, point to them, and have the child say them. This would logically (but maybe not strictly the whole-language way) be done after, or in connection with, helping the child know the sounds of the letters and letter-groups or phonemes. v Mark Thogmartin's chapter explains his ideas for teaching.
  In this initial stage of building a sight vocabulary, I would also help my child analyze the word by blending the letters left to right phonetically, discussing any variations from the expected such as in the diagraphs "ch" or "th" or other phonemes.
 

Children in US colonial times learned form wooden paddles similar to what is illustrated above. A printed sheet was protected by a layer of animal horn which was adequately transparent. The devices were called horn books.

How children recognize words

  Research we have discussed reveals that children learn new words better from story contexts than from traditional phonics, although we see a little of each method in the other. How do minds do so well at recognizing new words in print without learning a large set of letter-phoneme relationships? Phonics advocates tend to claim that these children are just memorizing the whole words, a seemingly impossible task given the number of words in a good vocabulary.
  Let's stop a minute here. Learning whole words without relating to the sounds of their letters, as was in vogue when I learned to read, is now condemned by both whole language advocates and phonics people. This is my excuse for being a not-so-good reader. But I lose at least part of it because my younger brother, whom I taught to read in sessions after I came home from school, is an excellent reader. Nice try. Well, maybe I'm a good teacher. Sorry, that's only a distraction. Let's get back to the topic.
  Until 1972, whole language people didn't have any good explanation for how good reading happened without traditional phonics. Then Donald MacKay9 studied speech errors resulting in words with parts replaced from other words in the sentence. For example, "Fix the big mixture" could become "Mix the big fixture." The brain hears "fix" as /f/+/ix/ and "mix" as /m/+/ix/. In this type of error, word parts tend to get hooked up in the wrong way. The slash marks ( / / ) here indicate sounds instead of printed words.
  MacKay's study led him to describe syllables in spoken words as heard into two parts: (1) any consonant sound that comes before the vowel sound, and (2) the vowel sound with any consonant sound that may come after it. The two parts are now known as onset and rime. These two sound parts for “fix” are /f/+/ix/.  If no consonant precedes the vowel, then the word has only a rime. Let's take a few more examples: "home school" = /h/+/ome/ /sch/+/ool/. (The h by itself in “home” doesn't seem like a significant word part until we compare with “Rome” which is /r/+/ome/.) "ice" is only /ice/ — no onset.
  The phonics breakdown for "school" would be /s/+/k/+/oo/+/l/. Studies verify that the brain tends to lump the phonemes into larger chunks, seeing words as onset and rime, and to make letter-sound correspondences using them.10
  Several studies have shown that children recognize new words better from analogous words than from their knowledge of phonics.11
  In summary, rather than seeing the letters in words as phonemes, children tend to group the letters into onsets and rimes. The rimes are actually conceptual groups of phonemes. When encountering new words in print, the mind remembers two parts of syllables (and, of course, whole syllables) in known words and judges the new word by parts from the old ones. According to the whole-language advocates, this means that children don't need to separate all the sounds and blend them into words before learning to read but they separate the sounds naturally as a result of reading.12
 

What happened in California?

  Unless you are familiar with the reading controversy, California might not seem relevant to reading. The state chooses educators who are experts in their curricular areas to set up standards for instruction. This is done every seventh year. When the standards are out, textbook publishers get busy preparing or revising their books because California decides what books its schools may choose from. Being the country’s largest buyer, they have the greatest influence. The following year the textbook lists are composed.
  In 1986 the group working on reading had been excited by the research showing the effectiveness of whole language reading methods and had established standards accordingly. Many citizens are advocates of traditional phonics, and Congress consequently asked Marilyn Adams, who works with a private research firm, to find out how to teach phonics better. Whether the method was effective was not the question and was not part of the support gathered to back the objective of her book, Beginning to Read.13 The research we have discussed about the difficulty in identifying phonemes was cited, and the goal became "phonemeic awareness."
  In California, the 1994-'95 Framework Commission reaffirmed the 1986 decision noting that teachers needed more preparation with the methods, which I assume may not have been followed carefully especially with the appearance of the "phonemic awareness" book.
  In 1995, the results of the recent national examinations (NAEP) was published. California was not at the top or even near the top, although children were reading. They were next to the bottom of the 39 participating states. No state should be near the bottom of such a list! Or so it seems. What could be the matter? Whole language, of course! The title of a Los Angeles Times report expressed the prevailing sentiment. "Dole Sees Problems in Schools and Blames Liberals."14
  Conclusions were not based on an examination of the situation.15 The following facts were ignored in defending the move to change reading instruction: 4 The size of classes in California was the largest or next to largest in the nation (depending on when they were measured). 4 The state's school libraries were among the poorest. 4 California’s Hispanic students scored the lowest of any ethnic group in any state. Their scores influenced the average. (Scores of California’s non-Hispanic groups matched those in other western states). 4 Twenty-five percent of the state’s population were new immigrants. 4 Only 2 percent of the teachers had been exposed to whole-language methods and sample lesson plans were only in preparation. (We may infer that some would have been teaching phonics and some would have let the children “read” without proper instruction.)
  California's new superintendent of public instruction authorized a Reading Task Force under the direction of a citizen who was prominent for promoting the phonics system. No reading professionals and not even any teachers below the fourth grade were on the committee. They recommended "an organized, explicit skills program that includes phonemic awareness, phonics, and decoding skills to address the needs of the emergent reader."
  In the meantime, in the cycle of their appointment, 110 professional educators established the list of recommended books, and the state board, in a rather unusual decision, promptly removed from the list books from two publishers of "predictable" books. Then in the middle of the cycle a new group of teachers was asked to write another framework. Next the legislature established a bill requiring books to include phonics, and teachers to be proficient in teaching it. Then the legislature acting through the fall of 1996, changed from their call for "balanced" reading instruction to a position of actively discouraging materials that enable children to figure out words in print in the natural way.
  This is the story of California. It is why the nation’s schools (and those in other English-speaking countries) have returned to a heavy emphasis on phonics.

The new phonics rationale

  Before learning most of what I have just shared, I found an article entitled, “Research-Based Reading Instruction: The Right Way.”16 It sounded good but was not what I expected. Because the author, Bill Honig, has clearly articulated the strong phonics position in language that appeals to educators, I’ll share some of his thoughts with a few comments.
  After describing the nation’s reading failures, Honig explains that the solution is to be found by looking at poor readers in the upper grades. He writes of “a powerful and persuasive consensus “ of educational, cognitive, and medical researchers, as well as our best teachers, “about the causes and cures of reading failure.” The more than 10,000 teachers he asked to describe them, report that they have “poor decoding skills.” And he adds that they struggle with too many individual words not knowing how to tackle new ones. They are also weak in vocabulary, spelling, strategic reading, etc. all because of reading failure.
  Honig points out that first grade decoding ability predicts 80 to 90% of reading comprehension in second and third grade and 40% in the ninth grade. “Predicts” is the correct word. We cannot say that good decoding causes good comprehension. Both poor decoding and poor comprehension must, to some degree, be caused by the same inadequate language ability. Reasons could be homes where English is not spoken or where children are not read to, or where children are not mature enough to begin school, etc. Also, we may assume that most of these children would not have learned the word-attack skills promoted in whole language programs. Likely lack of success with either type of program would lead to poor comprehension.
  According to Honig, “No method but full alphabetic screening of a word produces fast enough retrieval for the huge numbers of words in English. . . .  Context-driven decoding, even aided by partial alphabetic clues, is too slow and unreliable to serve as a fluent decoding tool.” Although this sounds logical, those who teach whole language carefully would disagree. The research we looked at and more that we haven’t, clarifies that kids can learn well with good whole language instruction. According to the NAEP test results, they learn better.

My opinion

  I would read to my kids letting them watch. We would talk about the stories. I would teach the alphabet and the sounds the letters usually represent and let them play with plastic letters or make letters with strings of clay. When they were ready I would let them see what words in the story were saying. I would explain common letter-combination phonemes and let them enjoy some phonics exercises. Then I would apply the principles we have discussed and which Mark Thogmartin explains in his chapter including stopping to help them sound out words. In brief, I would combine good principles from both methods.17 And as reading began, I would apply whatever method helped. You might do something different. Your kids will learn, too.

New, since the 7th edition went to press

   According to a brief article in U.S. News & World Report, Specialists from the National Research Council agree that children need both phonics and whole language instruction. 18

Teaching your beginner

  If your reader learns to stop often (maybe not always) to figure out new words using word-attack principles and confirmed by the context, reading skills will develop.
  Lack of success may not be from using the wrong methods. It could be a signal to wait for more maturity. Pushing is a temptation because you want the excitement of seeing your child learn to read. Use restraint. Pushing too soon or too hard could bring frustration to a child who is not ready. Unless your child is way past the typical reading age, don’t hold out the carrot saying, as soon as you can read, you can enjoy such-and-such.
  Whether with traditional phonics or a more casual approach, be diligent. You need to spend time with your child to help him or her avoid rushing over words which are not understood. Even with a word or two blocked out, most sentences can be understood correctly. (The principle is called “close.”) If your reader is depending on this, however, for meaning and not stopping to learn the words, carelessness develops and reading skill is in danger.
  In our teaching, we must avoid letting kids bypass the thought process by telling them a word before they have tried to figure it out or by letting them by with too many wild guesses losing their opportunity to understand the sentence. Neither should we hold them to instant perfection. We don’t in any other type of learning. In our own learning, we begin imperfectly and move to excellence.
  We must also avoid the idea that pronouncing strings of words is reading or that pronouncing individual words indicates knowing what they mean.
  Good readers have good habits of concentration and looking at the words. They also figure out, perhaps unconsciously, how letter combinations affect word sounds.

   1.  In 1955, Rudolph Flesch created a great stir by his book, Why Johnny Can’t Read. He correctly condemned the look-say instructional method of whole word memorization only, but his solution wasn’t, in my opinion, as good as it might have been.
   2.  McGuffey’s Eclectic Primer, Rev. ed., 1909, p. iii.
   3.  T. Clymer, 1963. “The Utility of Phonic Generalizations in the Primary grades.” The Reading Teacher 16: 252-258.
   4.  M. H. Bailey, 1967. “The Utility of Phonic Generalizations in Grades One Through Six.” The Reading Teacher  20: 413-418.
   5.  D. J. Bruce, 1964, “The Analysis of Word Sounds.” British Journal of Educational Psychology 34: 158-170.
   6.  Kenneth Goodman, A Linguistic Study of Cues and Miscues in Reading." Elementary English 42: 639-643, 1965. First graders correctly read, in the story, nearly two-thirds of the words missed on the list, second graders, 75%, and third graders 82%.
   7.  I. Mullins, J. Campbell, and A. Farstrup, 1993. NAEP 1992 Reading Report Card for the Nation and the States. National Center for Education Statistics. The score ratio for whole language / phonics rated as "little or no emphasis" was 218/222.
   8.  Jeanne Chall wrote: “a code-emphasis method — i.e. one that views beginning reading as essentially different from mature reading and emphasizes learning of the printed code for the spoken language — produces better results, at least up to the third grade.” Jeanne S. Chall, Learning to Read: the Great Debate. McGraw-Hill, New York, 1967, p. 307. The book was revised in 1983.
  In a 1988 article, Marie Carbo challenged Jeanne Chall’s conclusions that a code emphasis (bottom-up) is more effective than a meaning emphasis (top-down), and Chall responded in the same journal citing more studies that Carbo hadn’t mentioned. Then, in an article, “The ‘Great’ Debate – Can Both Carbo and Chall be right?”4 (Phi Delta Kappan, 1989, pp. 276-283.) Richard L. Turner reviewed the studies Chall had used in support of her ideas and found all but nine of them to have flawed methodology. From the few studies with a good research base, he concludes that systematic phonics is a little better than the whole-language approach until the beginning of the second grade, but after that, the difference disappears. I don't know why these results differ from those drawn from the NAEP scores. They would have been older studies.
  Also see my comments in Chapter 2 under “Public Schools” for research reports of general success of schools including a report on reading success in the US and New Zealand.
  Nancy Larrick, “De-trivializing our methods of reading instruction,” Phi Delta Kappan, November 1987, pp. 184-189. Children in one of the most difficult New York City school districts, many of whom didn’t hear English at home, showed remarkable success in reading. They were part of theSesame project which uses large numbers of ordinary children’s books instead of the typical basal readers. In this method (1) the teacher reads systematically to the children; (2) they are encouraged to read from many library books displayed in the room; (3) activities such as dramatization, discussion and illustrations make the books that are read to them more memorable; (4) key words (at the kindergarten level) are chosen by the children and displayed on cards around the room; and (5) the most interesting characteristic — phonics is not stressed. It is taught as children ask for help with their writing, and in quick crash courses before taking standardized tests.
  By late spring, 1987, all 350 first graders in theSesame project were reading in English, and all but three (who had been in this country less than six months) passed a comprehension test given by a school district evaluator. The project is based on an earlier research study comparing ten second-grade classes learning from this approach with ten more using conventional readers.
   9.  Donald D. MacKay, 1972. "The Structure of Words and Syllables: Evidence from Errors in Speech." Cognitive Psychology 3: 210-227.
  10.  R.E. Wylie and D.D. Durnell, 1970. "Teaching Vowels Through Phonograms." (The terms "onset" and "rime" had not yet been coined.) Also R. Treiman, 1983. "The Structure of Spoken Syllables: Evidence from Novel Word Games." Cognition 15: 49-84.
  11.  U. Goswami, 1986. "Children's Use of Analogy in Learning to Read. . . ." Journal of Experimental child Psychology 42: 73-83. U. Goswami, 1992 and F. Mead. 1992. "Onset and Rime Awareness and Analogies in Reading." Reading Research Quarterly 27: 150-162. M. Moustafa, 1990. An Interactive/Cognitive Model of the Acquisition of Graphophonemic Systems by Young Children. Ph.D. Dissertation, University of Southern California.
  12.  Margaret Moustafa, Beyond Traditional Phonics, Research Discoveries and Reading Instruction. Heineman, 361 Hanover St., Portsmouth, NH 03801-3912. 1997. p. 51. This book has been the souce for much of the material in this chapter. I've left you plenty to read.
  13.  Marilyn J. Adams. 1990. Beginning to Read: Thinking and Learning About Print. Cambridge, MA; MIT Press.
  14.  E. Chen and R.L. Colvin, The Los Angeles Times, July 18, 1996.
  15.  Mustafa, op cit. pp. 35, 36 She explains more problems with the condemnation of whole language.
  16.  Bill Honig, “Research-Based Reading Instruction: The Right Way,” Education Digest, Dec. 1997, pp. 15-22. Condensed from The School Administrator Sept. 1997. Honig, who is President of Consortium of Reading Excelence, in Emeryville, Calif., has also written Teaching our Children to Read: The Role of Skills in a Comprehensive Reading Program. Corwin Press.
  17.  An Illinois school was concerned that their whole language program needed more phonics. They found the combination successful. Kathy Batjes and Theresa Brown. ERIC Report ED409539. Improving Reading Achievement of First Grade Students by Integrating Phonics Skills into a Whole Language Curriculum. May 1997.
  Also, Barbara Matson. “Whole Language or Phonics? Teachers and Researchers Find the Middle Ground Most Fertile. The Great Reading Debate. Harvard Education Letter (v12 n2 p1-5 Mar.-Apr. 1996).
  18. U.S. News & World Report, "Two Good Ways to Teach Reading," March 30, 1998.