Showing posts with label Phonics. Show all posts
Showing posts with label Phonics. Show all posts

Sunday, April 10, 2016

Why Letter of the Week May Not Be Such a Good Idea

Teacher question:
Our district is trying to determine the proper pacing for introducing letter names/sounds in kindergarten. One letter per week seems too slow; 2 seems a bit fast. Most teachers are frustrated by 2 per week.

We are thinking about going with 1 for the first 9 weeks, then doubling up. This would have all letter names/sounds introduce by February. Can you offer some advise? How much is too much?

Shanahan response:
            This seems like a reasonable straightforward, simple question. And, it is, if you are a teacher, principal, or curriculum designer trying to plan a year of instruction. However, it is not the type of question that research takes on, so I can give you an answer, but it has to be one constructed on my understanding of the teaching of reading (research-based, but not research proven).

            The problem is that I could give a very specific answer like, teach one letter per week during kindergarten (and let’s face it, “Letter of the Week” is very popular). However, if I answered it in that way, I’d be ignoring some really important issues, like whether we want that much focus on individual letters and what is it that we want kids to know about letters.

            So let’s start with a really basic question:  What should a kindergartner know about this aspect of literacy by the end of the year? 

           In my opinion, kindergartners should know the names of all 52 upper and lower case letters. That means they should be able to name the letters presented to them in random order. They should also be familiar with one of the sounds associated with each of those letters—and it would be great if they knew both the “long” and “short” vowel sounds (so if I named or showed them a letter they could produce its sound, and if I made the sound, they could tell me the letter). Kindergartners should be able to sound out some one-syllable words or nonsense words using the letters they have learned. They should be able to fully segment single syllable words easily, and perhaps even be able to manipulate some of these sounds (adding them, deleting them, reversing them). And they should be able to print each of these letters and their names without having a visual model in front of them (and print their names).

            That description would be really easy to accomplish in some communities, where kids come to school already knowing letter names and some of the sounds, and it will be tougher in others. However, it would send kids off to Grade 1 ready to really become readers (especially if other aspects of literacy and language are being taught too).

            In any event, to accomplish all of this I would devote 30-45 minutes per day to these decoding issues—including the teaching of the letters (that's for full-day kindergarten--I would cut this in half in half-day situations). However, that does not mean you should sit kids down for 30-minute letter learning lessons—you might work on letters 2 or 3 times per day, for anywhere from 5- to 20-minutes per sitting.

            I think a combination of 1-2 letters per week is reasonable, but I wouldn't teach new letters every week. Remember letter naming or even letter sounding isn’t all that we want them to learn.

            For example, let’s say that on Week 1 I teach the “m” and “t” (letter names and sounds, upper and lower case), on week 2, the “p” and “h,” and on a third week, I teach only the letter “o” and its short sound. Then, on Week 4, there would be no new letters introduced. We would focus on using the 5 letters already taught. That means all of my decoding minutes would be spent on phonological awareness exercises focused on those specific sounds, blending various combinations of those letters (op, ot, om, top, tot, Tom, pop, pot, pom, hot, hop, etc.) into syllables, decoding and trying to spell syllables/words on the basis of the sounds alone. 

            If you gave each vowel its own week, and taught many, but not all, of the consonants in pairs, you could easily introduce all the letters over a single semester of kindergarten—and the students would have had at least 45 hours of practice with those letters; meaning a reasonably high degree of mastery should be accomplished by most kids.

            That means that those “non-letter introduction weeks”—like week 4 above—would be available 18 times during the year--fully half the year. You’d be spending as many weeks introducing letters as not introducing them. Those weeks would allow substantial amounts of phonemic awareness practice with those sounds, decoding work with those letters and sounds, invented spelling work and word construction with those letters and sounds, and ongoing review of all of that to ensure that the learning is really mastered.

            I would not save up those combination weeks until the second semester. I would salt them throughout the year to make sure that the learning was substantial and deep (meaning that kids would not just “know” those letters, but would be able to do something with them). Again, staying with my example above… 3 weeks of letter introduction, and then a week of consolidation might be followed by another week or two of letter introduction, and then back to consolidation with all the letters taught to that time, and that kind of a scheme could go on most of the year. Of course, if you noticed that your kids weren't retaining some of that, there would even be times that you could add in extra days or weeks of consolidation work as needed.

            With a plan like that, by summer, your kids would know their letters. But more importantly, they’d be able to perceive the sounds within words, and to engage in simple decoding and spelling using those letters and sounds. Outcomes not common in "letter of the week" teaching environments.

Monday, March 14, 2016

Putting One’s Underwear on First: Why Sequence Is Not Always So Important

 Teacher Question:  Is there a particular order in which teachers should teach the letter sounds?

  Shanahan responds:

            It makes sense to put your underwear on before you put on a skirt, shirt, blouse, or pant.

            Unless you’re Madonna.

            Then the usual ordering of things doesn’t necessarily get the job done. She changed the sequence from bra/blouse to blouse/bra and became a star. (It helped that she was wildly talented)

            Many teachers, principals, parents, and policymakers expect the proper ordering of letters and letter sounds in a curriculum to be more than a matter of convention or style, however. This question comes up often.

            It is hard explaining to them that there is no research-proven best sequence for teaching the ABCs or phonics. But that actually is the case.

            Back when the National Reading Panel report came out (2000), there was a similar hubbub in Congress. The Panel had found that phonics programs that had a clear sequence of instruction (‘systematic phonics”) were most successful. Consequently, officials wanted to require that everyone teach that sequence.

            The problem was that the Panel wasn’t touting a specific curricular ordering, just a planned curriculum. About 18 different phonics programs were examined in those studies, and they each had their own sequence of introducing letters and sounds. And they all worked.

            When it comes to teaching something like phonics, it makes sense to have a clear sequential system to follow. That way all the important patterns and elements get taught. As a classroom teacher, I tried to teach phonics in a more individual, diagnostic matter, keeping track of what I had covered with each child. It was an unholy nightmare; too much managing and less learning for the kids.

            That doesn’t mean the letter/sound orderings are completely arbitrary.

            For example, it makes good sense to offer earlier teaching of the letters and sounds that are used most often. Children seem to learn such letters—including the ones in their names—more quickly than the letters they don’t see much (Dunn-Rankin, 1978). That means one would be wise to teach letters like t, h, s, n, and the vowels, before taking on z, x, or k. One could learn these letters in any sequence, of course, but by teaching the most frequent ones early in the process, the teacher enables kids to read more words sooner—not a bad deal for them.

            When I was a becoming a teacher there was a controversy over whether to teach consonants or vowels first. Lots of argument, but not much data: Professors showed us that if you took all the vowels out of a message you could still read the text, so consonants were most useful; their argument for consonants. And others would argue back that no words are without vowels, so vowels clearly have higher frequencies.

            Common sense eventually won out. Instead of making it an all or none proposition, we figured out that by teaching a combination of consonants and vowels kids could read and write words earlier.
            Still another general guideline has to do with ambiguity. We should try to minimize confusion to make early reading easier. Whatever sequence is used should separate very similar letters. At one time, psychologists flirted with the idea of teaching highly similar letters together, so teachers could point out the differences, but studies found that it was better to go the separate route (Fang & Invernizzi, 2014). Don’t teach b and d together, or m and n. Letters that are visually or phonemically similar just need to be kept apart.

            Teach one of the confusables very well before introducing its partners. A student who already has strong purchase on either the /p/ or /b/ sounds, will have less trouble mastering the other. (Ws are confusing, not because of their great similarities with other letters, but because of the pronunciation of their names: I wish I had a nickel for every time I have told a young writer to sound out a w and he has said, “ Doooubbbblle-uuu…/d/”.)

            A related question has to do with capitals and lower case letters. Which of those do we teach first? Basically, lower case letters have greater value in reading. You simply see more of them, so the knowledge of such letters is more predictive of eventual reading achievement (Busch, 1980).

            Nevertheless, kids are more likely to come to school knowing their capitals (they are somewhat easier to teach because they are more distinctive, and so many preschool alphabet toys include capitals rather than lower case). And teaching them together is not a big deal (especially for the many that are miniature versions of the capital letters: c, k, m, o, p, s, v, w, x, y, z, for example).

            I guess what I’m saying is that the sequences of instruction for these elements are pretty arbitrary and you have a wide range of choices in how you do it. But that doesn’t mean that there aren’t commonsense guidelines that should shape these choices a bit. I definitely would not send my daughters to school with their underclothes on the outside, but then they aren’t Madonna.

Sunday, March 6, 2016

How Many Times Should They Copy the Spelling Words?

Two Teacher Asked Questions:

I have a question that was posed to me be an elementary principal. Her question was, "How many times does a student need to write a high frequency word before they feel secure with it?"  I must admit, I have never been asked this question before, and I cannot find research that addresses this specific question.  

The teachers in my school have kids copying missed spelling words 15 times. Is this a good idea?

            Everyone knows that, in order to accomplish great proficiency, musicians and athletes must engage in a great deal of repetitive practice. It would make sense that readers and writers would need to do the same thing to become accomplished with the words of their language.

            Yet analogies can be problematic. Two phenomena may be strikingly similar, but there are always differences (that’s why they are analogies). There definitely are useful drills in music and athletics, but that doesn’t mean that language works the same way.

            Repetition is important in language learning, but not necessarily the kind of repetition provided by writing a word over and over again.

            I have found no studies on the repetition of word writing or spelling, which surprised me. However, there is a substantial body of psychological research on word recognition (primarily because many psychologists have been interested in memory and word memories are relatively easy to study). None of these studies, as far as I can tell, look at comfort level; they are more likely to consider reaction times, correct responses, and generalization to other words.

            Basically, these studies suggest that the number of repetitions needed to learn a word is about 10-15 times, with lots of variation--among kids and words. For example, poor readers may require 12-25 reps to “learn” a word, while better readers may get away with only 8-12 (Lemoine, Levy, & Hutchison, 1993).

            Of course, words with regular sound-symbol relationships are learned about 25% faster than those with irregular spelling patterns. Studies also show that kids probably aren’t really memorizing words as much as they are becoming increasingly sensitive to intraword segments—combinations of letters within the words (Aaron, Wilczynski, & Keetay, 1998). This is probably why so many studies have found that words are much easier to learn through repetition than are nonwords (Jeffries, Frashish & Noble, 2009).

            Which points out why repetition may not be the best approach to “word memory.” Most of our word memories do not come through brute force memorization, though initially that is all we have. To learn words, people analyze the words for lexical and auditory/orthographic information, and these features are what allows later word recall for reading or spelling. Repetition helps—you can build some kind of word memory through rote repetition. But to develop a powerful, flexible understanding of words, you need to ask yourself, “repetition of what?"

            Kids are likely to learn a lot more words through pattern analysis (e.g., phonics) and the kinds of sorting activities recommended by Don Bear, Shane Templeton, and their colleagues.

            When I was a classroom teacher, I worried less about the number of repetitions kids made when they wrote, than I did about them trying to get them to build a visual memory of the words. I would have them “take a picture” with their eyes. Then I’d hide the word and ask, “Can you still see the picture?” Kids would then try to spell the word from memory.

            Given my success with that, I’m not surprised that the so-called, “Cover-Copy-Compare” method works (Joseph, Konrad, Coates, Vajcner, Eveleigh, & Fishley, 2001; Skinner, McLaughlin, & Logan, 1997). The kids visually analyze a word, then cover it up, try to recompose it from memory, and visually compare their written attempt with the word. That’s a form of repetition, but one with greater attention to building memory than to copying. Similar, the teacher who has kids trying to read or write a word repeatedly, might do better by supplementing or replacing this repetition with guided attention to the particular elements and features of the words (such as, "this 'e' is silent, or this one has two vowels side by side).

            Finally, research suggests that repetition may be important, but such repetition may better be built into language processing than working with lists and the like (Ideda & Morita, 2003). Reading or writing certain words again and again that are embedded in stories or articles, rather than presented list like, seems to provide greater support for learning (it may be a kind of interval training--you see the word, then you see some other words, then you see the word). Thus, writing high frequency words within the context of sentences or paragraphs may place more appropriate memory demands on learners.

          Kindergartners tend to have very limited word memory, while typical second- or third-graders often can learn new words very quickly. This is the result of developing knowledge of sound-symbol relationships and spelling patterns; that knowledge makes words "stickier." As Linnea Ehri has long pointed out, trying to learn a sight word when you are 5-years-old can be a major challenge—depending a lot on repetition and misleading mnemonics (like “monkey” has a tail). However, as kids gain an understanding of the spelling system, sight vocabulary develops much faster; often with very few exposures and no rote work at all.

          Repetition clearly has a role in reading and spelling--part of building that a system of word knowledge is getting some words into kids heads, but word analysis and repetition within natural language should be the major work horses in that endeavor. I encourage teachers to teach sight vocabulary to beginning readers, but I limit such memory work to about 5 minutes per day, and with both sight word learning and spelling, one tries to make such repetition sensible rather than rote. 

          It is true that high frequency words do not have typical spelling patterns, but it is rare that all of their elements are odd (e.g., the vowel pronunciation in "the" is a bit funky, but the /th/ is a more consistent element). Analyzing such words, rather than just repeating them again and again, is a better avenue to long-term learning than copying it over and over again.

Monday, December 7, 2015

Do We Teach Decoding in Small Groups or Whole Class?

Teacher question:  You are confusing me. You have said that we should “never do in small group what could have been done as well as whole class,” but you also say that phonological awareness and phonics instruction are more effective when they are taught in small group. What should be taught in small group and what can be taught in whole class?

Shanahan's response: 
            I’m a strong believer that when readers point out my contradictions that it is time to lather on plenty of Ralph Waldo Emerson. It was Emerson who famously said, “Foolish consistency is the hobgoblin of small minds.”

            That quote sure makes me feel better—and if the questioner feels like skulking away at this point, who could blame him?

            Okay, that’s not fair. You deserve an explanation of the seeming contradiction.

            First, the small group/whole class distinction. Small group instruction tends to be more effective than whole class teaching. In small groups, its easier for kids to stay focused, for teachers to notice error or inattention, and there is more opportunity for interaction and individual response.

            Although small groups may be more effective than whole class, that isn’t the real choice facing teachers. Their decisions must teeter between lots of whole class teaching versus small intermittent doses of small group teaching punctuated by independent seatwork. The kids may be advantaged by the small group work, but its benefits are balanced by the time spent on their own.

            Frankly, I see too much done in small groups. Teachers often present the same information over and over. If she has four comprehension groups, she’ll explain how to predict or when to summarize four times. My preference would be for the teacher to explain the strategy or skill to the whole class and then guide the student practice later in small groups. Similarly, when a teacher has two groups reading the same selection, I’d combine them, though it would make them larger.

            Now what about my phonics and phonological awareness statement?
That one is a little trickier. I was explaining the findings of meta-analyses of several studies that were aimed at determining whether phonological awareness or phonics instruction provided any advantage. The studies were comparing phonics teaching with little or no phonics teaching. And, over and over again explicit decoding instruction led to better reading performance.

            Although these were not studies that compared whole class versus small group phonics, the variation in studies made it possible for the meta-analyses to evaluate this feature. In some of the elementary studies the decoding skills were taught in small groups, while in other studies they were taught whole class. The conclusion was that studies that had looked at small group phonics teaching had bigger outcomes. Phonics and PA teaching work either way, but the small group delivery really magnifies those outcomes (perhaps because it facilitates the children seeing the teacher’s mouth movements and hearing the sounds clearly). (With children in preschool and kindergarten, such a comparison is not possible. All of the studies at these levels examined the outcome of decoding instruction delivered in small groups or individually.)

            However, lets not take such a finding as the final word. A program I know of here in Chicago, Reading in Motion, teaches such basic skills through the performing arts, engaging kids in songs and chants and so on.

            They deliver their engaging lessons whole class, but then follow up in small groups as necessary. If kids are making good progress from the whole class lessons, they don’t get small group work. If they struggle, the lessons are retaught in smaller groups to intensify it. Overall, this means less small group work than in many classes, but with higher rates of success. One can be both efficient and effective.

Seattle WERA Presentation Complex Text