Text Talk:

Capturing the Benefits of

Read-Aloud Experiences
for Young Children

By Isabel L. Beck, Ph.D.

Margaret G. McKeown, Ph.D.

Reprinted from The Reading Teacher

September 2001
W
hat should effective vocabulary instruction look like?
To answer this question, the National Reading Panel turned
to Dr. Isabel Beck and Dr. Margaret McKeown. The Text
Talk instructional program is a result of these top researchers’ years
of theoretical and practical inquiry into vocabulary learning.

Scholastic is proud to bring the proven-effective Text Talk instructional

program to classrooms across the country. Scholastic Text Talk follows
the instructional model designed by Dr. Beck and Dr. McKeown, as
described in this, The Reading Teacher reprint. Scholastic collaborated
closely with the authors to personally select the read-aloud titles, to
develop the instructional “talk,” and to detail the research-based
vocabulary instruction.

As offered by Scholastic, Text Talk is available in three levels for

grades K-3. Each level includes high-quality read-aloud trade books,
adhesive notes outlining teacher “talk” for the read-aloud, and lesson
plans for explicit vocabulary instruction that integrate essential
comprehension skills. Also included is a “Word Winner” chart, lending
library, and implementation DVD showing expert teachers modeling
Text Talk instructional strategies.

To purchase or find more information on Scholastic Text Talk, visit

www.scholastic.com/texttalk.
From Beck, Isabel L., & McKeown, Margaret G. (September 2001). Text talk: Capturing the benefits of read-aloud
experiences for young children. The Reading Teacher, 55(1), pp. 10–20. © 2001 International Reading Association.

Isabel L. Beck
Margaret G. McKeown

Text Talk: Capturing the

Keeping important text ideas in benefits of read-
focus and scaffolding children’s
responses and ideas are keys to aloud experiences
successful read-alouds.
for young children
What kind of texts?
C
oncern about young children’s lan-
guage development has recently cen- Texts that are effective for developing lan-
tered on the large individual differences guage and comprehension ability need to be
among children in vocabulary and comprehension conceptually challenging enough to require
abilities as they begin school (Biemiller, 1999; grappling with ideas and taking an active stance
Hart & Risley, 1995). The goal of the project we toward constructing meaning. The point is that
describe here is to enhance young children’s lan- young children can handle challenging content.
guage and comprehension abilities through in- Yet the limits of young children’s developing
depth and extensive experiences listening to and word recognition ability make it difficult to pro-
talking about stories read to them. vide challenging content in the books they read
Of course reading aloud to children has been on their own. However, because young chil-
pursued at home and in schools for centuries, and dren’s aural comprehension ability outstrips
indeed is probably the most highly recommend- their word recognition competence, challenging
ed activity for encouraging language and litera- content can be presented to young children from
cy (Adams, 1990; Anderson, Hiebert, Scott, & book selections that are read aloud.
Wilkinson, 1985; Goldfield & Snow, 1984). Yet
studies do not always show strong effects from What kind of talk?
reading aloud (Scarborough & Dobrich, 1994; Researchers suggest that the most valuable
Whitehurst et al., 1994). The issue at hand is to dis- aspect of the read-aloud activity is that it gives
cern what makes read-aloud experiences effective children experience with decontextualized lan-
for enhancing children’s language development. guage, requiring them to make sense of ideas
This article starts with consideration of what that are about something beyond the here and
the research literature suggests about the kinds now (Cochran-Smith, 1984; Heath, 1983; Snow,
of texts and kinds of talk that are most beneficial 1993; Snow & Dickinson, 1991; Snow, Tabors,
for read-aloud experiences. We then discuss Nicholson, & Kurland, 1995). As Donaldson
what we learned from observations of kinder- (1978) pointed out,
garten and first-grade teachers reading to stu-
dents. Next we provide an overview of Text children come to school well able to think and reason about
the world in situations that make human sense to them. What
Talk, an approach to read-alouds directed toward
they have to learn to do in school is to think and reason in
enhancing young children’s ability to construct “disembedded contexts”…to use symbol systems and deal
meaning. This section includes examples of with representations of the world. (pp. 88–89)
teacher/student interactions and suggests aspects
of reading aloud that need attention in order to The key to experiences with decontextual-
make these experiences more effective for chil- ized language that make them valuable for fu-
dren’s literacy development. ture literacy seems to lie in not merely listening

1 The Reading Teacher Vol. 55, No. 1 September 2001 ©2001 International Reading Association (pp. 10–20)
to book language, but in talking about the ideas. on major story ideas is not as simple as it may
Cochran-Smith (1984), Heath (1983), and Snow first appear. This is because young children tend
and her colleagues (Snow, 1993; Snow & to respond to stories by using what is easily ac-
Dickinson, 1991; Snow et al., 1995) all highlight cessible to them in contrast to the linguistic con-
the role of the talk that surrounds book reading tent (Neuman, 1990). Specifically, we observed
in becoming literate. According to Snow, quality how children frequently ignored text informa-
talk around books can promote familiarity with tion and responded to questions on the basis of
the pictures and their background knowledge.
relatively rare vocabulary, understanding the lexical and gram-
matical strategies for adjusting to a nonpresent audience,
The problem is that this reduces the opportuni-
identifying the perspective of the listener so as to provide suf- ties for children to construct meaning from de-
ficient background information, and knowing the genre- contextualized language which, as we have
specific rules for various forms of talk such as narrative and noted earlier, is essential for building mature lit-
explanation. (1993, p. 15) eracy skills.

Evidence for the role of talking about books What we learned from observations
in enhancing children’s language development Prevalence of responding on the basis of pic-
comes from studies by Dickinson and Tabors tures. Vivid, delightful pictures are a hallmark of
(1991), Freppon (1991), Morrow (1992), and children’s trade books, and children are naturally
Snow et al. (1995) who concluded, for example, drawn to them. However, if children rely on pic-
that “talk surrounding the text” (Morrow, p. 253) tures to construct their understanding of a story,
or “getting children to think about what was go- they may focus on characteristics of the pictures
ing on in the story” (Freppon, p. 144) were keys that interfere with constructing meaning of the
to literacy growth. story. As an example, consider our observation of
More specifically, Teale and Martinez a teacher reading a book called Socrates
(1996) concluded that the most effective talk in- (Bogaerts & Bogaerts, 1992) to a class of kinder-
volved encouraging children to focus on impor- gartners. The book’s cover shows a sweet-faced
tant story ideas and giving them opportunities dog wearing a pair of large red glasses. The
to reflect rather than expecting a quickly re- teacher indicated that the little dog was Socrates
trieved answer. Relatedly, Dickinson and Smith and that they would learn about him in the story,
(1994) found that talk that was “analytic in na- and began reading.
ture,” requiring children to reflect on story con- The story opens with a heart-wrenching de-
tent or language, was most beneficial. scription of Socrates’s parents being taken away
Dickinson and Smith’s (1994) and Teale and by the dog catcher and Socrates being left an or-
Martinez’s (1996) ideas about the most effec- phan with no friends and no one to care for him.
tive read-aloud strategies are quite consistent. After this opening, the teacher stopped and
The most effective features include focusing the asked, “What do we know about Socrates so
discussion on major story ideas, dealing with far?” The first response was, “He needs glasses.”
ideas as they are encountered in contrast to after Clearly the child ignored the linguistic content
the entire story has been read, and involving about Socrates’s situation and answered on the
children in the discussion with opportunities to basis of the picture. Thus this child constructed a
be reflective. However, it is clear from these in- completely different problem as the central focus
vestigations that the most effective read-aloud of the story.
strategies are far from the most common ones. In this case, the picture was not congruent
Our observations are consistent with those with the text content at that point. But even when
of Dickinson and Smith (1994) and Teale and pictures and text are congruent, it was our ob-
Martinez (1996). At the start of the project we servation that children often rely on the pictures
describe here, we observed kindergarten and for constructing meaning and thus miss oppor-
first-grade teachers reading to their classrooms tunities to engage in constructing meaning from
and found that they tended not to involve chil- the linguistic content.
dren in focusing on and discussing major story Children’s reliance on pictures is easy to ex-
ideas. Among the reasons this is the case is that plain, as pictures closely represent what children
in reading to young children, creating a focus are accustomed to encountering in the world

Text Talk: Capturing the benefits of read-aloud experiences 2

 around them. They can more readily derive infor- replied “Poison.” The teacher pressed, seeming
mation from pictures in comparison to text lan- to probe the student for reconsideration, “Did
guage. As Snow and Dickinson (1991) pointed out, she think it was poisoned?” Several students
comprehending and finding language to express replied “Yes,” and began to discuss incidents
ideas that go beyond the here and now is a new and they had heard about involving poisoned
challenging experience for young children. Halloween candy. This discussion took the class
Prevalence of responding on the basis of a considerable distance from the story ideas.
background knowledge. Research has shown that Children tend to report on their own experi-
background knowledge is a very important aspect ences because they can more readily derive in-
of understanding text (Anderson et al., 1985; formation from them in comparison to text
Beck, Omanson, & McKeown, 1982; McKeown, language. However, when what they report goes
Beck, Sinatra, & Loxterman, 1992; Pearson, too far afield, children can be distracted from the
Hansen, & Gordon, 1979). Attention to back- story or the inappropriate associations that they
ground knowledge has certainly found its way into bring in may be remembered as part of the story
instruction, to the extent that teachers customarily (Neuman, 1990; Nicholson & Imlach, 1981).
invite children to share background knowledge Teachers’ interactions in read-alouds. Thus
related to the story being read. The goal of invok- far a major point has been that several things
ing background knowledge is to integrate it with children “naturally” do in talking about a story—
text content in order to assist comprehension. But rely on the pictures and report their knowledge
in our observations we found that this is not nec- of things associated with the story—may stand
essarily the way background knowledge is used by in the way of their constructing meaning from
young children. Rather, we observed a tendency story information. Now let us turn to the
for children to respond to questions from back- teacher’s role in prompting children’s interac-
ground knowledge alone and ignore what had just
tions with text and the extent to which it supports
been read to them from the story.
constructing meaning.
For example, the story Curious George
In our observations of teachers reading to
Takes a Job (Rey, 1975) begins as follows, “This
children, two types of interactions seemed to
is George. He lived in the zoo. He was a good lit-
dominate. One was directed toward clarifying
tle monkey and always very curious. He wanted
to find out what was going on outside the zoo.” some content or unfamiliar vocabulary by ask-
At this point the teacher asked, “What do we ing, for example, “Does anybody know what a
know so far about George?” and the first re- ukulele is?” The other appeared to be attempts to
sponse was “He likes bananas.” Of course, be- involve children in the ongoing story by asking
cause George is a monkey he probably does like a question about what was just read. However,
bananas. But his fondness for bananas was not these questions were virtually always phrased
stated in the story, and more importantly it does in ways that produced only brief answers about
not help establish the major story concept of a detail. For example, “Harry likes everything
George as a zoo-bound monkey who wants to except taking a what?” “What kind of place
know what life is like in the world. were Mr. and Mrs. Mallard looking for to hatch
In some situations children simply took a their ducklings?” The problem with such ques-
notion from the text and drew an association to tions is that they constrain children’s responses
something in memory that was irrelevant or, at to a fact here and a detail there. Table 1 presents
best, tangential to the text situation. For exam- examples of questions we observed teachers ask
ple, in the story The Wolf’s Chicken Stew (Kasza, while reading Harry the Dirty Dog (Zion, 1984),
1987) a wolf leaves food for a chicken because The Mitten (Brett, 1989), and Brave Irene (Steig,
he’s trying to fatten her up for his dinner. 1986) and the children’s responses to those
Unbeknownst to the wolf, the chicken is feed- questions.
ing her large family with the food. As the story As can be seen, all the responses are correct,
moved to its climax the teacher began to probe and thus it is easy for a teacher to assume that
the children as to whether the chicken knows the understanding is in place. But dealing with these
source of the food. The teacher asked, “What did local issues does not add up to developing un-
she think the food was for?” and a student derstanding of a story.

3 The Reading Teacher Vol. 55, No. 1 September 2001

 Table 1
Examples of children’s responses to constrained questions

Questions Responses
As they started scrubbing, what came off ? Dirt
What does George want to do with his friend? Find him
How have things turned out for George, good or bad? Good
How is George doing at his job right now? Fine
George looks like he’s in a lot of what? Trouble
The mole found a new ____ Home
The mitten will be colored like snow, so it would be
hard to what? Find
Who needs the dress? The duchess

Text Talk overview most effective. That is, Questioning the Author
Our review of the research literature and our focuses on text ideas and encourages students’
observations in classrooms motivated the devel- participation in building meaning from those
opment of Text Talk, which is an approach to ideas as they read the text. Among the major dif-
read-alouds that is designed to enhance young ferences between Text Talk and Questioning the
children’s ability to construct meaning from de- Author is that the latter is directed to intermediate-
contexualized language. This goal includes not grade students who are reading their own texts in
only promoting comprehension, but also further- contrast to the focus of the current project, which
ing children’s language development. is reading aloud to kindergarten and first-grade
The project began with the selection of sto- children.
ries for kindergarten and first grade, the devel- The treatment of pictures and background
opment of questions for each story, and tryouts knowledge in Text Talk was influenced by our
of these materials. In the second phase we im- observations, discussed earlier. In Text Talk the
plemented Text Talk in kindergarten and first- pictures are for the most part shown after chil-
grade classrooms and worked closely with the dren have constructed meaning from what has
teachers to modify and augment the interactions been read. When background knowledge is elicit-
among the teacher, students, and text as issues ed, the teacher scaffolds children’s responses to
arose. The school was located in an urban pub- make clear the relationship of background knowl-
lic school district in a high-poverty area. edge to text ideas.
Seventy-five percent of the students received Beyond building comprehension of the spe-
free or reduced-cost lunch. All the students were cific story, Text Talk attends to children’s lan-
African American. guage development in two ways. One is that the
Text Talk interactions are based on open kind of questions asked elicit greater language
questions that the teacher poses during reading production. The other is that Text Talk takes ad-
that ask children to consider the ideas in the story vantage of some of the sophisticated vocabulary
and talk about and connect them as the story found in young children’s trade books by explic-
moves along. Our development of Text Talk was itly teaching and encouraging use of several
informed by our Questioning the Author work words from a story after the story has been read.
(Beck, McKeown, Hamilton, & Kucan, 1997). Table 2 provides an overview of components in-
Questioning the Author is an approach to text- volved in read-aloud experiences and a descrip-
based instruction that was developed around the tion of how those components are handled in
principle of “teaching for understanding.” Text Talk.
Features of Questioning the Author align very
closely with the features that Dickinson and Texts
Smith (1994) and Teale and Martinez (1996) Our criteria for selecting texts were that they
identified as making read-aloud interactions be intellectually challenging and provide the

Text Talk: Capturing the benefits of read-aloud experiences 4

 Table 2
How components of reading aloud are handled in Text Talk

Components Text Talk approach

Selection of texts Stories that exhibit an event structure and some complexities
of events to provide grist for children to build meaning.
Initial questions Interspersed open questions require children to describe
and explain text ideas, rather than recall and retrieve words
from text.
Follow-up questions Questions scaffold students’ thinking by using their initial
responses to form questions that encourage elaboration and
development of initial ideas.
Pictures In general, pictures are presented after children have heard
and responded to a section of text.
Background knowledge Invitations for background knowledge are issued judiciously
to support meaning building rather than encouraging stu-
dents to tap into tangential experiences.
Vocabulary Some sophisticated words are selected for direct attention
after reading and discussion of the story is completed.

grist for children to explore ideas and to use lan- Initial questions
guage to explain ideas. In particular, in choos- As noted above, we developed questions
ing stories we looked for some complexity of that teachers could use to initiate discussion at
events, subtleties in expressing ideas, or presen- important points in a story. In contrast to the
tation of unfamiliar ideas and topics. questions shown in Table 1, which constrain re-
Given our goal of promoting the construction sponses, questions developed for Text Talk
of meaning from linguistic content, we sought prompt students to talk about ideas. Table 3
books in which the linguistic content was shows examples of Text Talk questions and the
primary—that is, the book did not rely too heav- language they elicited from children collected
ily on the pictures for communicating the story. from pilot work in kindergarten and first-grade
A final criterion in consideration of construct- classes. As with Table 1, the examples are drawn
ing meaning was stories that exhibited an event from read-alouds of Harry the Dirty Dog (Zion,
structure rather than a series of situations, a for- 1984), The Mitten (Brett, 1989), and Brave Irene
mat that is sometimes used in books for young (Steig, 1986), albeit from different classrooms.
children. Examples of this format include
Seymour Simon’s Animal Fact/Animal Fable Follow-up questions
(1979), which presents one-page essays about It is important to emphasize that the kind of
different animals in response to a question (e.g., elaborated responses shown in Table 3 do not
arise automatically from asking open-ended
Do porcupines shoot their quills when they’re
questions. Indeed, in our Text Talk work, we
angry? Do goats eat tin cans?) and Family
found that children initially have difficulty con-
Pictures: Cuadros de Familia (Garza, 1990), structing these kinds of responses in contrast to
which presents a series of pictures with extended the customary responses of a word or two.
captions explaining “what’s going on” (e.g., Helping students to construct meaning requires
celebrating a feast day, harvesting oranges). teachers to take cues from a student’s initial re-
Although these are very attractive and interest- sponse, which for young children is often very
ing books for children to explore, they do not limited, and proceed from there. This territory
provide the extended, connected content for between a first, likely sparse response, and an
building meaning that is the focus of Text Talk. elaborated constructed response is the territory

5 The Reading Teacher Vol. 55, No. 1 September 2001

that requires teacher effort in creating thought- Class: Tree.
ful follow-up questions to support students’ con- T: Tree, OK. So what’s this all about?
struction of meaning. S: Monsters.
For example, consider a kindergarten class- T: What’s this story all about?
room in which the teacher was reading the story S: Giant.
Abiyoyo (Seeger, 1986). After the part of the story
in which Abiyoyo is introduced as “a giant called As the excerpt shows, the teacher made sev-
Abiyoyo…as tall as a tree and he could eat peo- eral attempts to get the children to expand their
ple up,” the teacher asked, “Who is Abiyoyo?” responses, but they did little beyond providing a
Her intent was that the children describe word or two. Even when the teacher’s question-
Abiyoyo, and understand why people fear him— ing moved beyond focusing on Abiyoyo himself
because he eats people. However, in the excerpt by asking “What’s this story all about?” children
below it is clear that children do not get very far stayed with simple one-word responses:
into these ideas. The discussion that ensued after “Monsters,” “Giant.”
the teacher’s initial question “Who is Abiyoyo?” Another kind of student response that re-
follows: quires consideration and careful follow up is re-
lated to a major theme of this article—children’s
S: A monster. difficulty interacting with decontextualized lan-
T: Did the story say he was a monster? guage. Consider, for example, a kindergarten
S: It’s a big green man. class read-aloud of The Giant Jam Sandwich
(Lord, 1972), a story about a town beset by a
T: A big green man. But does the story say what the big green
man was?
swarm of four million wasps. As the problem de-
velops, the villagers hold a meeting to discuss
S: He’s tall. how to solve their problem, but no one can come
S: A giant. up with a solution. At this point in the story, the
T: He’s a giant, and he’s tall as a _____ teacher asks, “What happened at the meeting?”

Table 3
Examples of children’s responses to open questions

Questions Responses

How does what Harry did fit in with what we already He doesn’t really want to get clean, he just wants to
know about him? stay dirty.
When the family looked out and said, “There’s a Because when he got all dirty, his family didn’t know
strange dog in the backyard,” why did they call who he was.
Harry a strange dog?
What’s Harry up to now? He decided to dig a hole and get the brush so he
could wash, and then they would recognize him.
They called Harry “this little doggie.” What does that That means that they don’t know that it’s their dog-
tell us? gie. They don’t know its name, so they just call him
little doggie.
Why do you think the children shouted, “Come quick”? Because the kids knowed that that’s the dog they had.
It says that “the mitten swelled and bulged, but Baba’s That it was strong, and she’s a good knitter.
good knitting held fast.” What does that mean?
What do you think Baba meant when she said, “If The gloves are the same color as the snow. That if
you drop one in the snow you’ll never find it”? you drop it in the snow it’s colored like snow.
It says, “Mrs. Bobbin…was tired and had a bad She’s sick, but she is still going to try and finish her
headache, but she still managed to sew the last stitch- dress.
es in the gown she was making.” What’s going on?

Text Talk: Capturing the benefits of read-aloud experiences 6

 but the children seemed unable to deal with the were being discussed. This approach to follow-
just-read linguistic event that the villagers could ing up children’s responses confirms findings
not come up with a solution to the wasp prob- from several studies. Orsolini and Pontecorvo
lem. The teacher calls on three different children (1992) found that 5-year-old children’s talk was
to answer the question, “What happened at the more likely extended when preceded by teacher
meeting?” but each child talks only about the repetition and rephrasing of what students had
general situation of the wasps in town: said. Nystrand and Gamoran (1991) found that
S1: There was the bees.
teachers who employed uptake—incorporating
previous student responses into subsequent
S2: Everyone was running around the town.
questions—had a strong positive effect on stu-
S3: They were stinging them. dents’ understanding of literature.
Another approach to following up children’s
The teacher asked the same question again, but initial responses included generic probes that
got a similar response: prompted them to explain: “What’s that all
S: Trying to sting them. about?” “What’s that mean?” We also found that
when children had difficulty responding to a
The teacher then reread the story portion about probe it was useful to reread the relevant por-
the meeting, with exaggerated expression, and tion of the text and repeat the initial question.
repeated, “What happened at the meeting?” The This helped students to focus on the text lan-
next child’s response was more related to the guage as the source for their responses. And
events of the meeting: even with all this, it takes time for students to ex-
S: He’s trying to get those things out of there so they don’t
pand their abilities to construct meaning from
sting. decontextualized language.

Children’s difficulty in responding to the Pictures

question likely occurred because it was much As noted earlier, during our observations we
easier for the children to respond to the general became very aware of how children often ig-
situation of wasps in town. That concept had al- nored the linguistic content and relied on pic-
ready been discussed, and is more vividly imag- tures to respond to questions about a story. Thus,
inable than a meeting of indecisive villagers. as we developed initial questions for Text Talk
Thus it was difficult for children to focus on and stories, we were alert to how children might use
respond to the meeting and its consequence. As the content of pictures. There were two situa-
such examples arose, we and the teachers began tions in which we deliberately decided to wait
to understand in a deeper way the difficulties to show pictures until after reading and discus-
children faced when asked to respond to decon- sion of a story portion.
textualized language, as much as the language One situation was when pictures mirrored
may have seemed explicit and clear to us as adult the linguistic content of a text. For example, in
readers. the story The Wolf’s Chicken Stew (Kasza, 1987),
As these examples illustrate, initial ques- after a wolf has been following a chicken, the text
tions may not bring forth meaningful responses reads: “The wolf crept closer. But just as he was
from young children. Yet simply asking more about to grab his prey….” The picture on this
questions will not necessarily prompt richer page shows the wolf on his hind feet about to
comments. Thus a great deal of our emphasis in pounce on the chicken. We wanted children to
working with teachers as they implemented Text talk about what was happening in the story at that
Talk was focused on how to follow up children’s point, so we posed the question “What’s happen-
initial responses in productive ways. Several ing?” If the children saw the picture as they were
concepts were developed that seemed useful. being asked that question, they certainly could ig-
One that was used frequently was to repeat and nore the linguistic content and respond just from
rephrase what children were saying. This both the visual. Because we wanted them to construct
encouraged more elaborated language and in- their idea from the text language, we did not
vited other children to connect to the ideas that show the picture until students had responded.

7 The Reading Teacher Vol. 55, No. 1 September 2001

 Another potential problematic situation with ing to children, teachers should use pictures ju-
pictures was when the content of pictures was diciously. Often this means after some event or
in conflict with what was going on in the text. idea has been explained linguistically.
For example, in The Bremen-Town Musicians
(Plume, 1980) there is a section in which a dog is Background knowledge
explaining to a donkey that he has run away be- As noted earlier, during our initial observa-
cause his master planned to shoot him. The don- tions we became aware of how often children re-
key then suggests that the dog join him, and they
sponded to questions about the story based on
leave for Bremen Town. The picture, however,
their background knowledge alone. As we worked
shows the dog hiding behind a tree and a man
with teachers in Text Talk, they too became cog-
with a rifle in his hand looking for him. This il-
lustration represents part of the story that the dog nizant of when children were using just back-
was relaying to the donkey. But the idea that ground knowledge rather than story information.
builds the plot of the story is that the dog and From these experiences teachers developed ways
the donkey have joined forces and are on their of acknowledging a student’s comment while
way to Bremen Town. The vividness of the pic- pointing out the distinctions between their own
ture could well lead children to misunderstand experiences and the story. For example, following
what was happening in the story at that point. the exchange about Curious George presented
Thus, we posed the question of “What’s going earlier, the teacher responded to the child by say-
on?” and elicited responses before showing the ing, “Monkeys do like bananas, but let’s think
picture. about what the story told us about George.”
Our observations of Text Talk showed us Additionally, consider the exchange from
that, for the children, the format of seeing the our earlier observation of The Wolf ’s Chicken
pictures later took some getting used to, but they Stew read-aloud, when children focused on the
soon came to understand the expectations of idea that the food left for the chicken might have
Text Talk and became more attentive to the lin- been poisoned. When we brought this example
guistic content as it was read. Several times to the teachers during a meeting about Text Talk,
when we observed teachers presenting Text Talk they had some suggestions for dealing with this
read-alouds early in the year, we noticed chil- type of situation. The tack they decided they
dren being caught off guard, unable to answer would take was as follows: “We sometimes do
the questions and asking to see the pictures. It hear about food being poisoned, especially bad
struck us that they had paid little or no attention people doing that at Halloween, but let’s think
to the words and were awaiting the pictures to about what’s happening in the story. Why did
fill them in on the story. When children were un- this food get left for the chicken? Who can re-
able to respond, the teachers would reread the mind us?” Presumably children would recall that
portion of text and explicitly remind children to the food was left to fatten up the chicken. From
listen to the words of the story to answer the
here the teacher could lead children to see that
question. With this support, children were able to
the food therefore would have been good food,
respond. As the implementation progressed, we
noticed that the teachers became alert to the im- not poisoned.
portance of timing for presenting pictures in or- Using the kind of exchanges noted above,
der to keep the linguistic content primary. With teachers helped children sort out the difference
this new awareness, they were surprised at the between simply responding from background
extent to which pictures were often the primary knowledge and responding from story informa-
source from which children answered questions. tion. Children need help in bringing background
The use of pictures needs to be considered knowledge to bear in appropriate ways, rather
from the perspective that constructing meaning than simply tapping into tangential experiences.
from text content is a major feature of what pre- There is evidence that readers’ elaborations of
pares one for becoming a successful reader. Thus knowledge and experiences that are not integrally
care needs to be taken that pictures do not cause related to text information can disrupt the process
students to skip attending to the language com- of comprehension rather than enhance it (Strang,
ponent of stories. That is, in the course of read- 1967; Trabasso & Suh, 1993).

Text Talk: Capturing the benefits of read-aloud experiences 8

 Vocabulary eat spinach” ask another child “What does it mean that [child’s
name] is reluctant to eat spinach?”
The acquisition of vocabulary is an obvious
focus for any program aiming to enhance chil-
Our previous program of work in vocabulary
dren’s literacy, because of the strong, well-
(see for example, Beck, Perfetti, & McKeown,
documented relationship that vocabulary has to
1982; McKeown, Beck, Omanson, & Perfetti,
reading proficiency in particular and school 1983) demonstrated the importance of maintain-
achievement in general (Anderson & Freebody, ing words after initial instruction. That is, if chil-
1981; Sternberg, 1987). Additionally, an impor- dren do not think about and use a word after initial
tant motivation for providing vocabulary experi- instruction it is unlikely to become part of their vo-
ences stems from the huge individual differences cabulary repertoire. Borrowing from our previ-
that exist in vocabulary size. In particular there ous work, especially the “Word Wizard” device
is an enormous discrepancy between high- and (where students earned points for seeing, hearing,
low-achieving learners (Graves & Slater, 1987; or using words they had been introduced to), we
Seashore & Eckerson, 1940). developed a simple way for teachers to encour-
Trade books are superb sources of vocabu- age and keep track of children’s awareness of in-
lary, and our Text Talk project takes advantage of structed words. Specifically, we created charts of
this by explicitly emphasizing vocabulary. From the words from each story, which enabled teachers
80 books (40 targeted to kindergarten and an- to tally each use or citing of a word. This appeared
other set of 40 targeted to first grade) we identi- to be quite successful, as each time we visited
fied about 1,500 words that could be taught to classrooms we noticed the continuing accumula-
children. A word was considered a good candi- tion of tallies next to words. Attention differed a
date if it seemed likely to be unfamiliar to young great deal according to individual words. For ex-
children but was a concept they could identify ample, the words nuisance and commotion were
with and use in normal conversation. We select- favorites among kindergartners, who often identi-
ed two to four words per story for direct teach- fied which of their classmates were being a nui-
ing following the story, and thus kindergarten sance and pointed out commotions in the
and first-grade children are provided instruc- classroom or hallway. Additionally, we observed
tion for approximately 100 fairly sophisticated and the teachers reported that often in story read-
words per grade. To illustrate the kind of words ing children recognized and remarked on the use
we included, Table 4 presents the words from of a word they had learned from an earlier story.
three of the stories used in Text Talk.
The instructional activities for each word be- Focus, monitor, and scaffold
gan by bringing to mind the use of the word from Enhancing young children’s comprehension
the story and explaining its meaning. Then stu- and language capabilities is essential for pro-
dents were involved with using or responding to moting literacy growth. Reading aloud and dis-
use of the word. Each activity also included hav- cussing what is read is an important avenue for
ing children repeat the word so they had a helping children deal with decontextualized lan-
phonological representation of what they were guage. But there are discrepancies between
learning. The following are teachers’ notes for common classroom practices in reading aloud
the activity for the word reluctant from A Pocket and those practices that have been found most
for Corduroy (Freeman, 1978). effective for laying the foundation for children’s
In the story, Lisa was reluctant to leave the laundromat without future literacy capabilities. Thus, in an effort to
Corduroy. Reluctant means you are not sure you want to do make reading aloud more beneficial for young
something. Say the word with me: reluctant. children, we developed Text Talk, an approach to
Someone might be reluctant to eat a food that they never had enhancing young children’s ability to build
before, or someone might be reluctant to ride a roller coaster meaning from text in which the teacher inter-
because it looks scary. sperses reading with open questions and discus-
Think about something you might be reluctant to do. Start sion, and follows each story with explicit
your sentence with “I might be reluctant to ______.” After attention to vocabulary.
each child responds call on another child to explain the re- From working with teachers as they imple-
sponse. For example, if a child says, “I might be reluctant to mented Text Talk, we can point to several con-

9 The Reading Teacher Vol. 55, No. 1 September 2001

 Table 4
Example of instructed vocabulary from three stories

Story Vocabulary

Abiyoyo (Seeger, 1986) disappear

precious
foolish

Alexander and the Wind-Up Mouse adventure

(Lionni, 1969) searched
envy

Amos & Boris (Steig, 1971) miserable

immense
leisurely

cepts that can guide the development of more ef- read-aloud discussion. Key to the task is keeping
fective read-aloud experiences. They include the important text ideas in focus while monitoring
following: children’s often limited responses and scaffold-
• awareness of the distinction between con- ing their ideas toward constructing meaning.
structing meaning of ideas in a text and
simply retrieving information from the text;
Beck teaches education courses and is a senior scientist at
• understanding the difficulty of the task
young children face in gaining meaning the University of Pittsburgh in Pennsylvania (647 LRDC,
from decontextualized language; 3939 O’Hara Street, Pittsburgh, PA 15260). McKeown is a
research scientist at the same university.
• designing questions that encourage chil-
dren to talk about and connect ideas and
developing follow-up questions that scaf-
References
fold, building meaning from those ideas;
Adams, M.J. (1990). Beginning to read. Cambridge: MIT Press.
• helping students to meaningfully incorpo- Anderson, R.C., & Freebody, P. (1981). Vocabulary knowledge.
rate their background knowledge and re- In J.T. Guthrie (Ed.), Comprehension and teaching: Research
duce the kind of surface association of reviews (pp. 77–117). Newark, DE: International Reading
Association.
knowledge that brings forth a hodgepodge Anderson, R.C., Hiebert, E.F., Scott, J.A., & Wilkinson, I.A.
of personal anecdotes; (1985). Becoming a nation of readers: The report of the
• awareness of how pictures can draw atten- Commission on Reading. Washington, DC: The National
Institute of Education.
tion away from processing the linguistic Beck, I.L., McKeown, M.G., Hamilton, R., & Kucan, L. (1997).
content in a text, and thus attention to the Questioning the Author: An approach for enhancing student
timing of the use of pictures; and engagement with text. Newark, DE: International Reading
Association.
• taking advantage of the sophisticated words Beck, I.L., Omanson, R.C., & McKeown, M.G. (1982). An in-
found in trade books by using them as a structional redesign of reading lessons: Effects on compre-
source of explicit vocabulary activities. hension. Reading Research Quarterly, 17, 462–481.
Beck, I.L., Perfetti, C.A., & McKeown, M.G. (1982). Effects of
Although reading a story to children is not a long-term vocabulary instruction on lexical access and reading
difficult task for a literate adult, taking advan- comprehension. Journal of Educational Psychology, 74,
506–521.
tage of the read-aloud experience to develop Biemiller, A. (1999). Language and reading success: From read-
children’s literacy is complex and demanding. ing research to practice (Vol. 5). Cambridge, MA: Brookline.
Even with awareness of what makes reading Cochran-Smith, M. (1984). The making of a reader. Norwood,
aloud most effective, it is difficult to keep dis- NJ: Ablex.
cussions consistently focused on the most pro- Dickinson, D.K., & Smith, M.W. (1994). Long-term effects of
preschool teachers’ book readings on low-income children’s
ductive features. Especially for young children, vocabulary and story comprehension. Reading Research
there is much to manage in conducting a good Quarterly, 29, 104–122.

Text Talk: Capturing the benefits of read-aloud experiences 10

 Dickinson, D.K., & Tabors, P.O. (1991). Early literacy: Linkages through social interaction (No. 61, pp. 11–24). San Francisco:
between home, school, and literacy achievement at age five. Jossey-Bass.
Journal of Research in Childhood Education, 6, 30–46. Snow, C.E., & Dickinson, D.K. (1991). Some skills that aren’t
Donaldson, M. (1978). Children’s minds. Glasgow, Scotland: basic in a new conception of literacy. In A. Purves & T.
Fontana/Collins. Jennings (Eds.), Literate systems and individual lives:
Freppon, P.A. (1991). Children’s concepts of the nature and pur- Perspectives on literacy and schooling (pp. 175–213).
pose of reading and writing in different instructional settings. Albany, NY: State University of New York Press.
Journal of Reading Behavior: A Journal of Literacy, 23, Snow, C.E., Tabors, P.O., Nicholson, P.A., & Kurland, B.F.
139–163. (1995). SHELL: Oral language and early literacy skills in
Goldfield, B.A., & Snow, C.E. (1984). Reading books with chil- kindergarten and first-grade children. Journal of Research in
dren: The mechanics of potential influence on children’s read- Childhood Education, 10, 37–47.
ing achievement. In J. Flood (Ed.), Promoting reading Sternberg, R.J. (1987). Most vocabulary is learned from context.
comprehension (pp. 204–215). Newark, DE: International In M.G. McKeown & M.E. Curtis (Eds.), The nature of vo-
Reading Association. cabulary acquisition (pp. 89–105). Hillsdale, NJ: Erlbaum.
Graves, M.F., & Slater, W.H. (1987, April). The development of Strang, R. (1967). Exploration of the reading process. Reading
reading vocabularies in rural disadvantaged students, inner- Research Quarterly, 2, 33–45.
city disadvantaged students, and middle-class suburban stu- Teale, W.H., & Martinez, M.G. (1996). Reading aloud to young
dents. Paper presented at the meeting of the American children: Teachers’ reading styles and kindergartners’ text
Educational Research Association, Washington, DC. comprehension. In C. Pontecorvo, M. Orsolini, B. Burge, &
Hart, B., & Risley, T. (1995). Meaningful differences. Baltimore: L.B. Resnick (Eds.), Children’s early text construction (pp.
Paul H. Brookes. 321–344). Mahwah, NJ: Erlbaum.
Heath, S.B. (1983). Ways with words. Cambridge, England: Trabasso, T., & Suh, S. (1993). Understanding text: Achieving ex-
Cambridge University Press. planatory coherence through on-line inferences and mental
McKeown, M.G., Beck, I.L., Omanson, R.C., & Perfetti, C.A. operations in working memory. Discourse Processes, 16,
(1983). The effects of long-term instruction on reading com- 3–34.
prehension: A replication. Journal of Reading Behavior, Whitehurst, G.J., Arnold, D.S., Epstein, J.N., Angell, A.L.,
15(1), 3–18. Smith, M., & Fischel, J. E. (1994). A picture book reading
McKeown, M.G., Beck, I.L., Sinatra, G.M., & Loxterman, J.A. intervention in day care and home for children from low-in-
(1992). The contribution of prior knowledge and coherent text come families. Developmental Psychology, 30, 679–689.
to comprehension. Reading Research Quarterly, 27, 79–93.
Morrow, L.M. (1992). The impact of a literature-based program
on literacy achievement, use of literature, and attitudes of chil- Children’s books cited
dren from minority backgrounds. Reading Research
Quarterly, 27, 250–275. Bogaerts, Rascal, & Bogaerts, Gert. (1992). Socrates. San
Neuman, S.B. (1990). Assessing inferencing strategies. In J. Francisco: Chronicle Books.
Zutell & S. McCormick (Eds.), Literacy theory and research Brett, Jan. (1989). The mitten. New York: G.P. Putnam’s Sons.
(pp. 267–274). Chicago: National Reading Conference. Freeman, Don. (1978). A pocket for Corduroy. New York: Puffin.
Nicholson, T., & Imlach, K. (1981). Where do their answers Garza, Carmen Lomas. (1990). Family pictures: Cuadros de fa-
come from? A study of the inferences which children make milia. San Francisco: Children’s Book Press.
when answering questions about narrative stories. Journal of Kasza, Keiko. (1987). The wolf’s chicken stew. New York:
Reading Behavior, 13, 111–129. Putnam & Grosset.
Nystrand, M., & Gamoran, A. (1991). Instructional discourse, Lionni, Leo. (1969). Alexander and the wind-up mouse. New
students’ engagement, and literature achievement. Research in York: Alfred A. Knopf.
the Teaching of English, 25, 261–290. Lord, John Vernon. (1972). The giant jam sandwich. New York:
Orsolini, M., & Pontecorvo, C. (1992). Children’s talk in class- Houghton Mifflin.
room discussions. Cognition and Instruction, 9, 113–136. Plume, Ilse. (1980). The Bremen-town musicians. New York:
Pearson, P.D., Hansen, J., & Gordon, C. (1979). The effect of Bantam.
background knowledge on young children’s comprehension Rey, H.A. (1975). Curious George takes a job. Boston: Houghton
of explicit and implicit information. Journal of Reading Mifflin.
Behavior, 11, 201–209. Seeger, Pete. (1986). Abiyoyo. New York: Aladdin.
Scarborough, H.S., & Dobrich, W. (1994). On the efficacy of Simon, Seymour. (1979). Animal fact/animal fable. New York:
reading to preschoolers. Developmental Review, 14, 245–302. Crown.
Seashore, R.H., & Eckerson, L.D. (1940). The measurement of Steig, William. (1971). Amos and Boris. New York: Farrar,
individual differences in general English vocabularies. Straus & Giroux.
Journal of Educational Psychology, 31, 14–38. Steig, William. (1986). Brave Irene. New York: Farrar, Straus
Snow, C.E. (1993). Families as social contexts for literacy de- & Giroux.
velopment. In C. Daiute (Ed.), The development of literacy Zion, Gene. (1984). Harry the dirty dog. New York: HarperCollins.

11 The Reading Teacher Vol. 55, No. 1 September 2001

About the Authors

Isabel Beck, Ph.D.

Dr. Beck has been acknowledged for
"bridging the gap between research and
practice" and is a member of the Reading
Hall of Fame, International Reading
Association. Dr. Beck has collaborated with
Dr. McKeown for over two decades to
conduct scientifically-based research on
effective vocabulary instruction.

Margaret McKeown, Ph.D.

Dr. McKeown is a leading researcher
examining the effect of vocabulary instruction
on reading comprehension. She has received
awards from the International Reading
Association and the Spencer Foundation
for applying theory and cognitive research
to practical classroom challenges.

Text Talk: Capturing the benefits of read-aloud experiences 12

 Scholastic Inc.
557 Broadway
New York, NY 10012

Visit us online at www.scholastic.com/texttalk

or call 1-800-SCHOLASTIC

Copyright © 2004 by Scholastic Inc. All rights reserved.

Item # 892052
10M 11/04