The prevailing assumption has been that a formal reading curriculum is inappropriate for kindergartners. However, district staff members and teachers in the Northern Lights School Division of Alberta were convinced that a "nurturing" approach to teaching reading would not endanger the children and might in fact prevent some of them from encountering academic difficulties in the primary grades and beyond.
WE'LL BEGIN with a simple proposition: let's teach our kindergarten students to read. We already know how to do it, so why don't we?
Within schools and school districts, decisions about curriculum and instruction in literacy have to be made on the basis of present knowledge and judgment. Such decisions can't wait until all controversies have been resolved and all the evidence is in with regard to available options. In the case of kindergarten, decisions about curriculum are complicated by debates about whether there should be a formal curriculum in reading or whether the components of the kindergarten program should be designed to develop the dimensions of emergent literacy only. But research on how to teach beginning readers grows apace, and we believe that we should take advantage of it.
In the Northern Lights School Division in Alberta, Canada -- a district of 20 schools and about 6,500 students -- we decided to design a formal reading curriculum for kindergarten, prepare the teachers to implement it, and conduct an action research study of student learning. Our decision stemmed from the judgment that research on beginning reading had reached the point where an effective, engaging, and multidimensional curriculum could be designed and implemented without placing our students at risk in the process. And if such a curriculum proved successful, it seemed likely that the much-publicized "learning gap" would be reduced.
Over the past five years in Northern Lights, we ("we" includes the superintendent, Ed Wittchen; the trustees; and representative teachers and administrators) had concentrated on the development of "safety nets" for low-achieving students at the second-grade level and in grades 4 through 12.(
The need for the safety net programs and our observation of the frustration and hopelessness experienced by students who needed help caused us to consider the K-3 literacy curricula and to explore whether we could strengthen them and so reduce the need for the later safety nets. We take seriously the statement by Connie Juel, who, in reacting to the National Research Council report Preventing Reading Difficulties in Young Children, wrote that "children who struggle in vain with reading in the first grade soon decide that they neither like nor want to read."(
For some decades, because of the concerns about not generating demands beyond the capabilities of the students or introducing students to reading in unpleasant ways, there has been a dearth of studies on formal reading programs for kindergarten. A few studies did suggest that formal reading programs in kindergarten could have positive effects that lasted throughout schooling.
For kindergarten interventions as such, though, we had to go back to Delores Durkin's work of 30 years ago. In building a kindergarten curriculum, we were not able to draw on a body of recent research on, say, alternative kindergarten reading programs or dimensions of learning to read at age 5. We drew on the literature relevant to learning to read in grades 1 through 3 and above. Building greater literacy is a matter of considerable importance, and not damaging our students is of even greater importance. But it may be that the concerns about hurting students are based on images of brutal and primitive curricula rather than on humane and sophisticated approaches. Certainly those concerns are not based on reports of failed attempts.(
We made the decision that there would be no danger to the students if we proceeded deliberately and, particularly, if the teachers tracked the responses of the children carefully and were prepared to back off or change their approach if a student appeared to be stressed. Not to challenge students cognitively might be an even larger mistake than challenging them. In addition, we wanted the early experience to be not only effective but joyful -- learning to read should be a delightful experience.
Our view of a nurturant curriculum appears to differ widely from the image that many people have of a reading curriculum for young children, and we believe it is that image that causes them to shy away from formal literacy instruction for kindergartners. We did not imagine students with workbooks, alphabet flash cards, or letter-by-letter phonics drills. Instead, we imagined an environment in which students would progress from their developed listening/speaking vocabularies to the reading of words, sentences, and longer text that they had created, where they would examine simple books in a relaxed atmosphere, where they would begin to write with scribbling and simple illustrations, where they would be read to regularly, and where comprehension strategies would be modeled for them through the reading and study of charming fiction and nonfiction books. If the work of childhood is play, we imagined the students playfully working their way into literacy.
Pathways to Literacy: Designing the Curriculum
Our idea for a nurturing curriculum came from developments in the field of curriculum having to do with several of the emergent literacy processes. Most of the literature in this area presents ideas about and studies of students in grades 1 through 6. We saw this literature as defining dimensions for early literacy that could be incorporated into components of a kindergarten curriculum. Essentially, we categorized dozens of studies around the several dimensions:
• The development of sight vocabulary from the students' listening/speaking vocabulary and the study of words encountered through wide reading.(
• The need for wide reading at the developed level. At the beginning, students can engage at the picture level and, gradually, can deal with books at the caption level as they learn how meaning is conveyed by the authors.(
• The regular study of word patterns, including spelling.(
• The need for regular (several times daily) writing and the study of writing.(
• The study of comprehension strategies.(
• The study, by both teacher and students, of weekly and monthly progress, including the levels of books the students can read, sight words learned, phonetic and structural analysis skills, information learned, and fluency in writing.(
For our early literacy curriculum, we found that the Picture Word Inductive Model -- derived from the tradition of "language experience" with the addition of concept formation and attainment models of teaching -- was very important. The core of the language experience approach is the use of the students' developed listening/speaking vocabulary.(
The Picture Word Inductive Model, as the name suggests, begins with photographs of scenes whose content is within the ability of the students to describe. For example, the photographs might show aspects of the local community. The students take turns identifying objects and actions in the picture. The teacher spells the words, drawing lines from the words to the elements in the picture to which they refer and so creates a picture dictionary. The students are given copies of the words, and they identify them using the picture dictionary. They proceed to classify the words, noting their similarities and differences. The teacher then selects some of the categories for extended study. Both phonetic features and structural characteristics are studied. The teacher models the creation of titles and sentences, and the students create some of their own by dictating them and learning to read the dictations. In the same fashion, the teacher creates paragraphs, and the students gradually learn to assemble titles and sentences into paragraphs about the content of the picture. The picture word cycles (inquiries into the pictures) generally take from three to five weeks.
A major assumption underpinning this view of the curriculum is that students need to become inquirers into language, seeking to build their sight vocabularies and studying the characteristics of those words as they build generalizations about phonetic and structural characteristics.
The curriculum was designed to facilitate growth through each of its strands -- building vocabulary, classifying, creating sentences and paragraphs, and reading -- in an integrated fashion so that each strand will support the others. As indicated above, as sight words are learned, phonetic and structural concepts will be developed through the analysis of those words. Similarly, the construction of sentences and paragraphs will be related to the sight vocabularies that are being developed. As the children read, they will identify known words and attack new ones through the phonetic, structural, and comprehension skills they are developing.
Providing Staff Development to Support Implementation
Once we decided that such a curriculum was feasible, designing staff development was the next step. We needed a program that was oriented to help the teachers both implement the curriculum and become a positive learning community that would study student learning and take pleasure in colleagueship and inquiry. Eight teachers in three schools in the Grand Centre/Cold Lake area were involved in the initial effort. The school faculties had agreed formally to participate, and all eight kindergarten teachers had agreed as well. Two had taught reading in the primary grades in the past, but none had attempted a formal literacy curriculum in the kindergarten. Two were first-year teachers. The superintendent, cabinet, and board of trustees were supportive, and meetings explaining the curriculum were held with parents in the spring and early fall.
The staff development included demonstrations, the study of early literacy, the analysis of practice, and the study of student learning, following the format developed by Bruce Joyce and Beverly Showers.(
The Action Research Inquiry
For the action research component of the initiative, the eight teachers and the district staff members were asked to focus on two questions: Did the multidimensional curriculum work? Did the students learn to read and to what degree, including the extent of their comfort with the process and their feelings about reading?
Informal observation was important, but the teachers were also provided with tools for the formal study of the students' learning of the alphabet, acquisition of vocabulary, general language development (including phonemic awareness), books studied or read, and development of the competence to manage unfamiliar books, including extended text, using the procedure developed by Thomas Gunning.(
To what extent is the variance in achievement explained by gender, by developed language competence as students entered kindergarten, and by class group -- variables that occur repeatedly in the literature and are reported as factors in many studies? In the first year, all 141 kindergarten-age students in the three schools were enrolled and were included in the study. In all three schools, students came from a considerable variety of socioeconomic levels, and some 15 students came from First Nations reservations. Teacher judgment indicated that just one of the children entered kindergarten reading at any level. Just one student could recognize all the letters of the alphabet (tested outside the context of words).
Throughout the year the data were collected, summarized, and interpreted with respect to the response of the students. Here we concentrate on the most salient aspects of the students' learning. All eight kindergarten classes followed similar patterns. Differences between the classes were small by comparison to the general effects. For us, this was very important. Had it been that only half of the teachers had been able to implement the curriculum successfully, we would have had to do some heavy thinking.
Recognition of letters of the alphabet. In early October, the mean number of letters recognized (out of 52 upper- and lower-case letters) was 31. In January, the mean was 46. In March, it reached 52. That is, all the students could recognize all the letters out of context. Letter recognition was associated with the acquisition of sight vocabulary, but one was not necessarily a function of the other. The learning of sight vocabulary appeared to pull letter recognition as much as the learning of the letters facilitated the acquisition of sight vocabulary.
Acquisition of sight vocabulary. Our inquiry focused both on how many words were being learned and on the students' ability to learn new words. The learning of words was studied in terms of the Picture Word cycles, which ranged from about four to six weeks in length. Both the number of words learned in the cycles and the increased efficiency developed by the students were of interest here. The data below are taken from one of the classes.
Cycle 1. Twenty-two words were "shaken out" of the picture. At the end of the first week, the average number of words identified in an out-of-context assessment was five. By the end of the fourth week, the average was 16, and one student knew all 22.
Cycle 2. Twenty-two words were shaken out. At the end of the first week, the average number that the students could identify out of context was 12, and by the end of the third week, the average number identified was 20.
Cycle 3. Twenty-eight words were shaken out. At the end of the first week, the mean number of words recognized out of context was 20, and at the end of the second week, the mean was 26, with just three students recognizing 24 and none recognizing fewer than 24.
All the students appeared to increase in efficiency so that, by the end of January, they were able to add to their sight vocabularies, within the first week or two, just about all the words shaken out of the picture. For all sections, the mean percentage of words recognized after two weeks of the first cycle was 30%. By the third cycle, the mean for two weeks had risen to 90%.
Retention of words. In May, random samples of six students in each class were tested with respect to out-of-context recognition of the words that had been shaken out through the year -- for example, about 120 words in the class cited above. Mean retention was 110. In addition, words added through the generation of titles, sentences, and paragraphs were learned, many of them in the high-frequency "useful little words" category. In the class used as an example, those additional words added up to over 100.
Had the students had difficulty developing a sight vocabulary or retaining it, we would have had a serious warning signal. But such a signal did not develop, and, more important, the increase in capability was a positive signal. By midwinter, the students were mastering words within two weeks that had taken them four or five weeks in the first cycle.
Classification of words. Once the words were shaken out, they were entered into the computer, and sets of words were given to each student. (The students could examine them and, if they did not recognize one, could use the picture dictionary to identify it.) Classifying the words was an important activity. The students were asked to sort their word cards according to the characteristics of the words. The teachers modeled classifications of various types throughout the year. In the first cycles, most students built categories on the presence of one or more letters. Later, more complex categories emerged. The teachers selected categories for instructional emphasis and led the students to develop new words and unlock unfamiliar words by using the categories. For example, having dealt with work, works, worked, worker, and working, the students could hunt for other words from which derivatives could be made. Or, knowing work and encountering working in their reading, they could try to unlock it as they learned how the -ing suffix operates.
The teachers studied the categories that students were developing, keeping an eye on the phonetic and structural principles that were emerging. The results are too complex to summarize briefly, but, on the whole, about 30 phonetic and about 20 structural concepts were explored intensively.
Transition to reading books. Throughout the year, a profusion of books was available to the students. Books were carried home for "reading to and with," and little books generated from the Picture Word activities went home to be read to parents. As the students began to learn to read independently, books at their levels accompanied them home. Our records show that 80% of the students encountered 50 or more books in this fashion, in addition to any books from home or libraries.
The assessment of independent reading levels was built around the Gunning framework, in which the students attempt to read unfamiliar books at the following levels:
• Picture Level: single words on a page are illustrated.
• Caption Level: phrases or sentences, most but not all illustrated.
• Easy Sight Level: longer and more complex, mostly high-frequency words.
• Beginning Reading: four levels, progressively longer passages, and less repetition and predictability.
• Grade 2A: requires good-sized sight vocabulary and well-developed word-attack skills.
When an assessment is administered, students read aloud books at each level, beginning with the simplest, and their deviations from print are noted. They are asked comprehension questions after the book has been read. Reaching fluency with total comprehension places a student at a particular level.
In the December assessment, all the students were able to deal with books at the Picture Level, and about one-fourth could manage Caption Level books comfortably. By February, about one-fourth had progressed to the Easy Sight Level, and a handful could manage books at a higher level.
Once again, had the students not been able to approach any level of text competently, we would have had a warning that our curriculum was failing. However, the children were progressing beyond the reading of the sentences and paragraphs developed in each Picture Word cycle and were beginning to be able to manage simple books "almost independently."
In June, the independent test team administered the assessment using a specially assembled set of books from United Kingdom publishers to reduce the likelihood that the books would be familiar to the students. The aggregated results for the eight classes were indeed encouraging:
Level Percentage Reaching Level
Picture 2
Caption 26
Sight 30
Above Sight 42
All eight classes apparently succeeded in bringing all the students to some level of print literacy. About 40% of the students appeared to be able to read extended text, and another 30% manifested emergent ability to read extended text. Indeed, 20% reached the Grade 2A level, which includes long and complex passages and requires the exercise of complex skills both to decode and to infer word meanings. All the students could manage at least the simplest level of books.
We felt it was very important that there were no students who had experienced abject failure. Even the student who enters first grade reading independently at the picture level is armed with skills in alphabet recognition, possesses a substantial storehouse of sight words, and owns an array of phonetic and structural concepts. However, a half dozen students will need to be watched closely because, even if they were able to handle books at the caption level, they labored at the task, manifesting difficulty either in recognizing relationships between text and graphics or in using their phonetic or structural generalizations to attack unfamiliar words.
We studied the data to determine whether gender or socioeconomic status influenced levels of success, and they did not. The distributions of levels for boys and girls were almost identical, as were the distributions for students having or not having subsidized lunches.
Typically, in our district, about 20 kindergarten students would have been referred as having special needs in those eight schools. At the end of this year, just two students were referred, both for speech problems.
Comfort and satisfaction. During the year, parents voiced their opinions regularly, and in May we prepared simple questionnaires for both the parents and the children. We asked the parents a series of questions about the progress of their children and whether they and the children believed they were developing satisfactorily. The children were asked only whether they were learning to read and how they felt about their progress. We were trying to determine whether there was any discomfort that we were not detecting. But in response to our survey, no student or parent manifested discomfort or dissatisfaction related to the curriculum. However, some parents were anxious at the beginning and remained worried at the end of the year. Some were concerned that we had not taken a "letter by letter" synthetic phonics approach and worried that future problems might develop as a consequence. But even these parents appeared to believe that their children were progressing well "so far."
A Year Later: Leaving First Grade
Throughout first grade, we followed the students, and, at the end of the year, we gave them the Gray Oral Reading Test,(
In June 2003, 47 students, a randomly selected half of the 94 students still enrolled in the district, were administered the Gray Oral as they exited grade 2. Their average GLE was 5.0 (the national average of exiting grade-4 students). The distributions of male and female scores were almost identical. Five students (10%) scored below the average of exiting second-grade students. Typically, 30% of the students in this district or nationally in the U.S. and Canada do so.
In subsequent years, we will continue to monitor the progress of the students from each year, and we will follow the lowest-achieving students most intensively.
Interpretation
The problem that faced us was whether research on beginning literacy had reached the point that we could design multidimensional curricula to introduce young children to reading with comfort and satisfaction. In our efforts to learn how much an initiative in kindergarten curriculum might improve literacy learning, reduce the likelihood of failure by students thought to be at risk, and also benefit students not thought to be at risk, our first experience must be described as positive. We will follow the students through the grades, and we will continue to scrutinize the curriculum.
The teachers were all new to a formal kindergarten reading curriculum. In the first year, they were scrambling to master a considerable number of unfamiliar instructional models, particularly the Picture Word Inductive Model, and they spent considerable energy tracking the progress of the students and trying to figure out whether they were proceeding optimally and whether the tasks were well matched to them. With greater experience, they will no doubt provide many ideas for improvement.
The issues of "developmental readiness" become moot if the knowledge base permits us to design effective and humane kindergarten curricula in reading. The progress of the students in these eight classes equals the progress of students in average first-grade classrooms and surpasses it in one very important way: no children failed, whereas one-third of the students in average first grades usually do. The half-dozen students who gained the least nonetheless arrived at first grade with substantial knowledge and skill.
In the next few years, we'll learn how these students do in the upper elementary grades, where similar efforts to change the curriculum are under way. Thus far, our results have been encouraging, but there are 400 students to follow now. We certainly want to continue the outstanding achievement we have seen so far, but we also hope to close the door on poor achievement and eliminate the need for the safety net programs. We'll see. Right now, our hypothesis is that a strong, multidimensional, formal reading program for kindergarten students can change the picture of achievement in the primary grades. Moreover, 5-year-old children, given a strong and humane curriculum, can learn to read at least as well as first-graders usually do, but without the high failure rates of so many first-grade classrooms.
We hope that our Northern Lights teachers, and all others in every venue, will set high standards and also treat their students affirmatively. We are bothered when states, provinces, and districts set goals at such a low level that they expect that 2% or 3% of the students will creep up to the next level of achievement in any given year. Ninety-five percent is a better goal. Nearly all of our little second-grade graduates can now read with the best of upper-elementary-grade students. So could nearly all of the students in all school systems.
By Bruce Joyce; Marilyn Hrycauk and Emily Calhoun, with the Northern Lights Kindergarten Teachers
BRUCE JOYCE is director of Booksend Laboratories, St. Simons Island, Ga. MARILYN HRYCAUK is director of instruction in the Northern Lights School Division #69, Alberta and EMILY CALHOUN is director of the Phoenix Alliance, St. Simons Island, Ga. The Northern Lights Kindergarten Teachers are BEV GARIEPY, CHRISTINE REYNOLDS, MELANIE MALAYNEY, CAROL KRUGER, JENNIFER LAWTON-GODZIUK, ELAINE BLADES, CHRISTINE CAIRNS, ANDREA FAMA, and GLORIA LANE.
