Skip to content
ascd logo

Log in to Witsby: ASCD’s Next-Generation Professional Learning and Credentialing Platform
April 1, 2003
Vol. 60
No. 7

Walking with Rosie: A Cautionary Tale of Early Reading Instruction

Schools that focus entirely on teaching decoding skills in the early grades neglect the essential vocabulary knowledge that students need to become competent readers.

Walking with Rosie: A Cautionary Tale of Early Reading Instruction - thumbnail
In an urban kindergarten classroom, the teacher reads Rosie's Walk(Hutchins, 1986) to her students. In the story, Rosie the hen goes for a walk, oblivious to the sneaky fox following her. Along the way, Rosie walks around a pond, a mill, and a haystack.
The teacher has printed the words pond, mill, and haystack on large cards, which she places on the floor in front of her students. As she rereads the story, she points to the word cards and asks the students to walk around them the way Rosie walks around each of the locations in the book. The class discusses the meanings of the words pond, mill and haystack.
Then the teacher gathers the students in a circle, placing the card with the word pond on it in the middle. The teacher guides the students in stretching out the sounds in the word as they pronounce it, and she models a gesture for grabbing the last sound in the word, /d/. She has students point to the letter d that ends the word pond as they say the ending sound. She does the same with mill and haystack.
The teacher in this classroom was one of 13 kindergarten teachers we observed as we followed the literacy development of nearly 200 students from preschool through 1st grade. We found the Rosie's Walk lesson particularly impressive because it focused students' attention on multiple aspects of words. First, the teacher used the context of the book to introduce students to the meanings of the words pond, mill, and haystack. Although these words seem relatively simple, many young children in urban communities have not been exposed to them. The teacher then asked students to focus on the ending consonant sounds of each word, building their phonological awareness. Having students match these sounds to the letters in the words helps them learn letter-sound relationships.
Such lessons encourage students to develop both their comprehension skills, which build on vocabulary knowledge, and their decoding skills, which require letter-sound knowledge. By using a book to contextualize the word meanings and using word cards to show the spelling of the words, the teacher gives students the opportunity to connect each word's meaning, sounds, and spelling. We call this approach anchored word instruction.
This lesson's integrated approach to early literacy instruction differs from the teaching that we observed in many classrooms. Most of the time, we saw teachers focusing primarily on phonological awareness and decoding skills without paying attention to the meanings of the words.
In an example typical of much of the instruction we saw, a teacher in one classroom places a set of objects on the floor in front of her students. She names the objects: toothpaste
turtle, teapot, toothbrush, tank, cotton, cat, cap, car, and cookie. The teacher guides the students to sort the objects into a /t/ pile and a /k/ pile on the basis of their beginning consonant sounds. She also asks the students to identify the letter that goes with each beginning sound. This lesson promotes students' letter-sound knowledge but does not focus on teaching them vocabulary or comprehension.

The Need to Improve Vocabulary Skills

Many schools focus on teaching students about sounds in words and the letters associated with them because research has found that these skills have powerful instructional effects on developing word recognition skills (Adams, 1990; Ehri, 1998; National Reading Panel, 2000; Snow, Burns, & Griffin, 1998). But the ability to decode—although crucial for reading success—is not sufficient (Baker, Simmons, & Kameenui, 1998; Cunningham & Stanovich, 1998; Juel, Griffith, & Gough, 1986). Oral vocabulary knowledge also plays a pivotal role in reading comprehension (Dickinson & Tabors, 2001; Hart & Risley, 1995).
Many students may appear to be progressing well according to reading assessments in the early grades because tests at this level have relatively simple words and content. But when these students reach the later grades, their lack of vocabulary knowledge becomes increasingly apparent (Chall, Jacobs, & Baldwin, 1990; Cunningham & Stanovich, 1997; Juel, 1988). After students learn to decode, further growth of their reading comprehension depends on their knowledge of word meanings. Vocabulary knowledge and other comprehension skills that students should acquire while they learn to decode become crucial to understanding what they read (Hoover & Gough, 1990). If this store of knowledge is not well developed, students might experience comprehension problems later in school.
Lack of word knowledge can have a direct impact on the development of reading comprehension because of the kinds of words that commonly appear in books. “Book language” draws on vocabulary that occurs relatively rarely in oral conversations. For example, a writer may use the word leap for jump, or thump for tap (Snow, 1983). If students do not understand the nuances of more sophisticated “book language,” their comprehension will certainly suffer.
Students of low socioeconomic status, like the students in our study, are overrepresented in the group of students with weak vocabulary knowledge. Research suggests that students from low socioeconomic backgrounds know around 6,000 fewer words than their middle-class peers do when they start school (Hart & Risley, 1995). This gap seems to increase over time.
But the picture is not entirely bleak. Research on vocabulary instruction suggests that teaching vocabulary skillfully can increase students' reading comprehension (Stahl & Fairbanks, 1986). But the reading field's current preoccupation with decoding skills and its lack of attention to teaching vocabulary and listening comprehension in the early grades may result in an imbalance in instruction and seriously compromise students' vocabulary development.

Classroom Activities and Student Language Arts Development

Our work with a longitudinal low-income sample of almost 200 early readers brought our attention to this potential problem. The students in our study attended six different preschools that fed into three elementary schools in contiguous neighborhoods in a southeastern city. Sixty-eight percent of the students qualified for free or reduced-price lunch. Sixty percent were African American, 35 percent were white, and 5 percent were of mixed or other ethnic origins. Fifty-one percent were male and 49 percent female. By studying the word reading and vocabulary knowledge of these students in the context of the instruction they received, we began to see patterns in how teaching influences literacy skills.
We began by evaluating students' development in decoding and vocabulary from preschool through 1st grade using the Woodcock Diagnostic Reading Battery (1997). Figure 1 shows that the students in our sample demonstrated a relative weakness compared with national norms in both letter-sound identification and oral vocabulary when they began preschool, as do many students living in poor neighborhoods (Hart & Risley, 1995; Whitehurst & Lonigan, 1998). Each year, our students' decoding skills improved. By the middle of 1st grade, their average decoding skill score was slightly higher than the national norms. This score represented considerable growth over two school years.
Figure 1. Students in the Study Compared with the National Norming Sample

Walking with Rosie: A Cautionary Tale of Early Reading Instruction - table 1

Letter-Sound Identification

Oral Vocabulary

Mean of Our SampleMean of Norming SampleMean of Our SampleMean of Norming Sample
1st Grade10310089100
Average scores on the Woodcock Diagnostic Reading Battery.
Our assessments of vocabulary development through 1st grade, however, painted a less promising picture. The gap between the students in our sample and national norms remained the same over time. Although the students made gains in oral vocabulary from year to year, they did not catch up overall to their more advantaged peers.
To understand the relationship between the students' development in literacy skills and the kinds of instruction they received in kindergarten and 1st grade, observers from our research team visited these 13 kindergarten and 1st grade classrooms every three weeks during language arts instruction, which typically lasted 90 minutes. On a laptop computer, the observers recorded what they saw and heard in the classrooms and coded their narratives for the types of literacy activities that predominated during the instructional time. These painstakingly detailed observations allowed us to estimate how much of each type of activity an individual student experienced during kindergarten and 1st grade.

Language Arts Activities Observed

The students' impressive improvement in word recognition skill reflected the concerted effort of their schools to use careful, targeted, and differentiated instruction to promote decoding ability. The schools focused on these skills through professional development in phonological awareness and word study training. Most teachers adapted their instruction to address students' varying needs through flexible reading groups in kindergarten and 1st grade, and they individualized letter-sound instruction and word study to each group's needs.
Teachers evaluated the students' decoding skills regularly and used this information to reorganize the groups so that students would work with other students according to their developmental level in reading skill. The teachers did not use prepackaged, scripted instructional programs. Although the teachers had access to basal readers, they rarely used them. Instead, they created such materials as chart-sized poems and teacher-made word cards to accompany leveled readers.
Teachers used a systematic approach to phonics instruction, focusing on beginning consonants first, followed by ending sounds and middle vowel sounds, but this approach was informed by the teachers' understanding of students' development in literacy acquisition rather than a teacher's manual. In line with this developmental sequence, teachers showed differences in the focus of their instruction from kindergarten to 1st grade.
Focus in kindergarten. Figure 2 shows the proportions of dominant language arts instruction devoted to each activity on average and the wide variation we observed among different classrooms. Because a major goal of kindergarten is to introduce students to the alphabet and because students in this study entered school with relatively weak letter knowledge, it is not surprising that letter-sound activities accounted for nearly one-third (29 percent) of the dominant language arts instruction. Teachers used phonics, rhyming exercises, discussions of word families, and activities in which students matched letters with words to foster letter-sound knowledge.
Figure 2. Activities in Kindergarten Classrooms

Walking with Rosie: A Cautionary Tale of Early Reading Instruction - table 2




Oral Language24%0%46%
Anchored Word12%0%38%
Percent of dominant language arts activities observed in 13 classrooms (198 students).
The teachers varied greatly, however, in the extent to which they incorporated letter-sound activities into their instruction. Our observers noted that at least one teacher did not emphasize letter-sound relationships at all; other teachers devoted as much as 75 percent of their dominant language arts instruction to these activities.
The next most common kindergarten activity, focusing on oral language development, accounted for about one-fourth (24 percent) of the dominant language arts instruction. Teachers addressed oral language through a variety of techniques, including discussions of books read aloud to students, with brief detours to discuss the meanings of some words. As in our analysis of letter-sound instruction, we found wide differences in the extent of dominant oral language-related activities in kindergarten. In fact, observers noted that some students did no oral language-related activities, whereas others were involved in them up to 46 percent of the time.
Reading activities—for example, independent reading, choral reading, and round-robin reading—dominated almost as often as oral language activities: 23 percent on average. Again, we saw much variation among teachers in reading activities.
Writing activities, which included both free writing and instruction in spelling, grammar, and punctuation, accounted for 12 percent of dominant instructional time in kindergarten.
What we call anchored word instruction—the type of instruction illustrated by the kindergarten teacher in our opening example of Rosie's Walk—dominated 12 percent of language arts instructional time in kindergarten. Such instruction generally involved calling attention to a word from a book by printing it on a word card. Although teachers devoted instructional attention to the letter-sound relationships in the words, they also discussed the meanings of the words in the context of the text so that the lesson provided multiple anchors to help students learn about the word: meaning, spelling, and sounds.
Focus in 1st grade. Language arts activities in 1st grade changed to reflect the students' developing skills as readers. As Figure 3 shows, the proportions of dominant language arts activities again varied greatly among students. Letter-sound activities were no longer the most frequent instructional activity on average; the instructional emphasis shifted to reading and discussing books. This shift in instructional focus makes sense; by 1st grade, most students have developed stronger decoding skills and no longer need as much practice linking letters to sounds.
Figure 3. Activities in 1st Grade Classrooms

Walking with Rosie: A Cautionary Tale of Early Reading Instruction - table 3




Oral Language31%0%63%
Anchored Word7%0%37%
Percent of dominant language arts activities observed in 13 classrooms (198 students). Note: 2% of activities did not fit into the coding scheme.
Oral language activities became the most frequent type of dominant language arts activity in 1st grade (31 percent). The next most prevalent activity was reading (28 percent), followed by letter-sound activities (19 percent). The relative amount of writing (13 percent) stayed about the same in 1st grade as in kindergarten, but anchored word instruction declined slightly to 7 percent.

Instruction That Promotes Vocabulary Development

In contrast to these students' growth in decoding skills, their vocabulary skills remained quite low from preschool through 1st grade, nearly one standard deviation below the national average each year. To determine what kinds of instructional techniques were most effective in developing students' vocabulary skills, we used a multiple regression model that took into account the students' oral vocabulary in the previous years. We examined how students' scores on the Woodcock Oral Vocabulary test could be predicted on the basis of the amount of exposure they received to various instructional activities in kindergarten and 1st grade.
We had thought that oral discussions during read-alouds of children's books, for example, would positively affect students' growth in oral vocabulary. We were wrong. Neither reading children's books aloud to students nor mentioning vocabulary during these readings significantly predicted oral vocabulary during kindergarten or 1st grade. This finding, unfortunately, resonates with other studies of the effects of instruction on vocabulary growth (Stahl & Fairbanks, 1986). As suggested by Biemiller (1999) and Beck, McKeown, and Kucan (2002), incidental instruction may not be substantive enough to significantly boost the oral language development of students who enter school with weak oral vocabularies.
We also found that letter-sound instruction was slightly, but significantly, associated with increased oral vocabulary in kindergarten, although not in 1st grade. For this young age level, perhaps isolating words and attending to some aspects of the letters and sounds in those words has a positive influence on oral vocabulary growth.
We found relationships in 1st grade between performance in oral vocabulary and two forms of instruction associated with word recognition. First, 1st graders who received more anchored word instruction tended to score higher in oral vocabulary. Figure 4 shows the relationship between the percentage of dominant anchored word instruction and oral vocabulary scores. The three lines represent groups of students who had low, average, and high levels of oral vocabulary in kindergarten relative to their peers in the study. The more anchored word instruction they received, the higher they are predicted to score in oral vocabulary in 1st grade.

Figure 4. Anchored Word Study Activities and Students' Oral Vocabulary

In contrast, more instruction in letter-sound relationships was associated with lower vocabulary scores. In particular, the more a student experienced unanchored letter-sound instruction—in which the focus remained on the letter sounds rather than explicit references to word meanings or contextual links (as in the previous example of sorting objects)—the lower the student's predicted oral vocabulary scores (see fig. 5).

Figure 5. Letter-Sound Activities and Students' Oral Vocabulary


Implications for Early Reading Instruction

Because we did not design this study as an experiment, we cannot interpret these results causally. Nevertheless, the implications deserve consideration.
First, teachers in the study effectively delivered instruction that fostered students' word reading skills. The teachers' strategies included emphasizing word study in kindergarten and delivering differentiated instruction to small groups of students.
Second, oral language exposure, such as that provided in reading children's books aloud, does not seem in itself to significantly promote students' vocabulary growth in the early grades. Perhaps teachers need to improve the oral language activities that they implement in their classrooms during story reading to help students capitalize on the rich language in books and to boost students' vocabularies.
Third, attention to the letters and sounds of words in kindergarten appears to foster students' vocabulary growth. Attending to words as specific units of language to be analyzed might promote beginning readers' early word awareness and, consequently, their oral language development.
When teachers carry over this emphasis on letter-sound instruction into 1st grade, however, it may have detrimental effects on oral vocabulary progress. Perhaps when teachers focus instruction solely on reading words rather than on making meaning of text, they pay less attention to building students' vocabularies. This possibility does not mean that schools should curtail letter-sound instruction. Rather, teachers should augment such necessary and effective instruction with more substantive attention to vocabulary and comprehension in the early grades.
Although we observed it relatively rarely, anchored word instruction may have beneficial effects on oral language development. Highlighting several important aspects of a word—meaning, spelling, and sound—provides students with a wealth of information that helps to anchor that word in their memories. This multifaceted approach to word learning, incorporating instruction in both decoding and vocabulary, may prove useful in students' development of oral language. We are currently implementing an experimental study to investigate this possibility.
Ultimately, effective early reading instruction must help students learn to identify words and know their meanings. With so much research emphasizing the importance of early development in both word reading and language skills, we must consider how to provide instruction that fosters students' vocabulary development without losing the promising results of effective instruction in decoding. It does little good, after all, to be able to sound out the words pond, mill, and haystack if you have no idea what they mean.

Adams, M. (1990). Beginning to read: Thinking and learning about print. Cambridge, MA: MIT Press.

Baker, S., Simmons, D. S., & Kameenui, E. J. (1998). Vocabulary acquisition: Research bases. In D. S. Simmons & E. J. Kameenui (Eds.), What reading research tells us about children with diverse learning needs (pp. 183–217). Mahwah, NJ: Erlbaum.

Beck, I., McKeown, M., & Kucan, L. (2002). Bringing words to life: Robust vocabulary instruction. New York: Guilford.

Biemiller, A. (1999). Language and reading success (Vol. 5). Cambridge, MA: Brookline.

Chall, J., Jacobs, V., & Baldwin, L. E. (1990). The reading crisis: Why poor children fall behind. Cambridge, MA: Harvard University Press.

Cunningham, A., & Stanovich, K. (1997). Early reading acquisition and its relation to reading experience and ability 10 years later. Developmental Psychology, 33, 934–945.

Cunningham, A., & Stanovich, K. (1998, Spring/Summer). What reading does for the mind. American Educator, 1–8.

Dickinson, D., & Tabors, P. (2001). Beginning literacy with language. Baltimore: Brookes.

Ehri, L. (1998). Grapheme-phoneme knowledge is essential for learning to read words in English. In J. Metsala & L. Ehri (Eds.), Word recognition in beginning literacy (pp. 3–40). Mahwah, NJ: Erlbaum.

Hart, B., & Risley, T. (1995). Meaningful differences. Baltimore: Brookes.

Hoover, W., & Gough, P. (1990). The simple view of reading. Reading and Writing, 2, 127–160.

Hutchins, P. (1986). Rosie's walk. New York: Macmillan.

Juel, C. (1988). Learning to read and write: A longitudinal study of 54 children from first through fourth grades. Journal of Educational Psychology, 80(4), 437–447.

Juel, C., Griffith, P., & Gough, P. (1986). Acquisition of literacy: A longitudinal study of children in first and second grade. Journal of Educational Psychology, 78, 437–447.

National Reading Panel. (2000). Report of the National Reading Panel: Teaching children to read. Washington, DC: National Institute for Child Health and Human Development.

Snow, C. (1983). Literacy and language: Relationships during the preschool years. Harvard Educational Review, 53(2), 165–189.

Snow, C., Burns, S., & Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Research Council.

Stahl, A., & Fairbanks, M. (1986). The effects of vocabulary instruction: A model-based meta-analysis. Review of Educational Research, 56, 72–110.

Whitehurst, G., & Lonigan, C. (1998). Child development and early literacy. Child Development, 69(3), 848–872.

Woodcock, R. (1997). Woodcock Diagnostic Reading Battery. Itasca, IL: Riverside.

Connie Juel has contributed to Educational Leadership.

Learn More

ASCD is a community dedicated to educators' professional growth and well-being.

Let us help you put your vision into action.
From our issue
Product cover image 103033.jpg
The First Years of School
Go To Publication