FIRST GRADERS' SOCIALLY CONSTRUCTED DEFINITIONS OF READING
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
To my family--
Bill, Mom, Dad, Cilla, and David
I am grateful to a number of people who lived the graduate school
and dissertation experiences with me. Without these people I might
still have made it through, but the process would have been even more
arduous and much less rewarding.
First, I would like to thank the members of my committee. My
chairperson, Dr. Ruthellen Crews, is a special kind of teacher who is
always eager to learn something new. Her unfailing interest in the
study helped me maintain my own enthusiasm. Dr. Crews has also been
my good friend. I have often been revived and energized by our long,
wide-ranging discussions and the laughter we shared in her office.
Dr. Dorene Ross has always willingly put her own work aside to answer
my questions, help me clarify my thoughts, and point me in the right
direction. Her high standards and quick mind have challenged me to be
the best I can be. Dr. Rod Webb, who gave me a whole new way to think
about schooling, has been an inspiration to me. I have appreciated
his classes, his book, his careful critiques of my papers, his en-
couragement, and his unfading smile. Dr. Ted Hipple is the reason I
stayed on for a Ph.D. I am grateful for his confidence in me, his
sense of humor, and even his "snittish editorial comments." I thank
these talented people for their help and their friendship.
To the teacher who opened her classroom to me I am deeply grate-
ful. Amidst the constant activity and demands of her busy days, she
always greeted me with a smile and eagerly made time to help me in any
way she could. Her warmth and enthusiasm made my task easier. I have
appreciated her trust, her friendship, and all she has taught me about
working with young children. In addition, I thank her students, who
shared themselves with me, and the many teachers at the school who made
me feel so welcome there.
I am also indebted to Amos Hatch, who, more than anyone, made
graduate school the productive adventure it has been. His confidence
in me helped me to have more confidence in myself. I am forever grate-
ful for the many hours we spent thinking, talking, writing, and laughing
together. Other friends who have been especially important to me
include Dr. Sue Kinzer and Cherry Kay Travis. I thank Dr. Kinzer for
always having time, in spite of an often hectic schedule. I thank
Cherry Kay for technical assistance above and beyond the call of duty.
My parents, Anne and Gene Bondy, have helped me in many ways.
Their love and support have provided a firm foundation from which I
could continue to grow, learn, and make mistakes. They have always let
me know that they believed in me and that what I was doing was important.
My husband, Bill Dunn, who has been working on a Ph.D. of his own,
deserves a medal for surviving my degree. These have been difficult
years, but Bill has been patient and supportive throughout. His love
and understanding have given me strength and have brought us closer
TABLE OF CONTENTS
ACKNOWLEDGMENTS . .
ABSTRACT . .
I BACKGROUND FOR THE STUDY
Statement of the Problem
Significance of the Study.
Definition of Terms. .
Design of the Study. .
Scope of the Study .
Review of the Literature .
Studies of Children's
. . .
. . .
. . .
. . .
. . .
. . .
Perceptions of Reading:
Cognitive-Developmental Perspective .
Studies of Children's Perceptions of Reading:
Social-Interaction Perspective .
II METHODOLOGY . . .
The Setting . .
Selection of Research Site .
Gaining Entry to the Site .
Description of the Site .
Research Methods and Procedures .
Overview . .
Asking Ethnographic Questions .
Collecting Ethnographic Data .
Participant observation .
Interviewing . .
Unobtrusive measures .
Making an Ethnographic Record .
Analyzing Ethnographic Data .
Researcher Qualifications and Biases.
Validity . .
III CHILDREN'S DEFINITIONS OF READING .
Reading Is Saying Words Correctly .
Reading Is Schoolwork . .
Reading Is a Source of Status .
Reading Is a Way to Learn Things .
Reading Is a Private Pleasure. .
Reading Is a Social Activity .
IV CHILDREN'S DEFINITIONS OF READING: PRODUCTS
INTERACTIVE PROCESS. . .
Children's Entering Views of Reading .
Low Group . .
High Group. . .
Teachers' Practices ...........
Reading Is Saying Words Correctly .
Reading Is Schoolwork . .
Reading Is a Source of Status .
Reading Is a Way to Learn Things. .
Reading Is a Private Pleasure and a Soc
V CONCLUSIONS AND IMPLICATIONS .
Relationship of Findings to Previous Studies
Use of Findings to Research Community. .
Use of Findings to Practitioners .
A PROJECT OUTLINE. . .
B OCTOBER INTERVIEW WITH MRS. SAUNDERS .
C NOVEMBER INTERVIEW WITH MRS. SAUNDERS .
D DECEMBER INTERVIEW WITH MRS. SAUNDERS. ..
E INTERVIEW WITH TEACHER IN ADJOINING ROOM .
F INTERVIEW WITH KINDERGARTEN TEACHERS .
REFERENCES. . . ... ..
BIOGRAPHICAL SKETCH . .
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
FIRST GRADERS' SOCIALLY CONSTRUCTED DEFINITIONS OF READING
Chairperson: Dr. Ruthellen Crews
Major Department: Curriculum and Instruction
The purpose of this study was to investigate in detail the defini-
tions of reading constructed by children in one first-grade classroom.
The researcher assumed a social-interaction perspective by which
definitions of reading were viewed as meanings individuals assigned to
reading as a result of their interactions in social contexts. The study
focused on two guiding questions:
1. What are the definitions of reading constructed by members
of the low and high ability reading groups in one first-grade
2. Are there patterns in children's definitions within and
across ability groups?
Qualitative research methods were used to collect and analyze data.
Observations were conducted throughout the school day for 150 hours
during the first four months of school. These observations focused on
children's speech messages about reading, their reading-related be-
havior, and their use of reading materials. Formal and informal inter-
views were conducted with children in the low and high reading groups,
their teacher, and the children's kindergarten teachers. In addition,
children's cumulative school records were examined.
Data analysis was an ongoing process which proceeded through several
phases. The analysis revealed six definitions of reading:
1. Reading is saying words correctly.
2. Reading is schoolwork.
3. Reading is a source of status.
4. Reading is a way to learn things.
5. Reading is a private pleasure.
6. Reading is a social activity.
Although definitions were not clearly differentiated by group, low
group children tended to construct the first three definitions, and high
group children tended to construct the second three definitions. No
definitions were shared by all children, and most children used more
than one definition to guide their reading-related behavior. Definition
construction was found to be the result of an interactive process
between the children and the teacher. Specifically, the variables
which seemed to be related to children's definitions were cognitive
developmental factors, children's entering views of reading, home
experiences with written language, personality factors, and the context
in which the defining process took place.
The study highlighted the complexity of teaching and learning
processes. The results suggested that in order for teachers to pro-
vide effective reading instruction for all students, they must become
sensitive to the students' ways of thinking about reading.
BACKGROUND FOR THE STUDY
Recent observational studies indicate that prepackaged commercial
programs seem to define the nature of reading instruction in many
elementary classrooms (Duffy & McIntyre, 1980, 1982; Durkin, 1978-1979;
Goodlad, 1983). Researchers have uncovered a pattern of instructional
practice in which teachers move students through basal materials, assign
workbook pages, listen to students recite from texts and workbooks, and
respond to their answers. This input-output model of teaching in which
teachers or materials deposit information and students are expected to
display a particular response fails to account for what may be a criti-
cal factor in teaching and learning: the learners' perceptions of the
object of instruction. That is, learners' perceptions of the skills,
processes, or materials to be learned may influence the way they pro-
cess instruction as well as their learning outcomes. In the area of
reading, children's perceptions of the nature, purposes, and functions
of reading may have great bearing on their progress as readers.1
Statement of the Problem
If children came to classrooms as blank slates, teachers' jobs
would be much easier. However, children enter school with well-formulated
In this study the notion of "perceptions" is interpreted differently
from the traditional cognitive interpretation. See the discussion on
pages 5-7 for clarification.
perceptions of the world in general and classroom life in particular.
As they interact with objects and individuals in their environment,
new perceptions develop and old ones are confirmed, modified, or re-
jected. Children use these perceptions in a continuous struggle to
make sense of their world, and their perceptions influence their inter-
pretation of and their response to classroom events. As long as chil-
dren's perceptions and teachers' perceptions are synonymous, classroom
life proceeds smoothly. However, a disparity in views may interfere
with teaching and learning processes. If teachers are to provide
effective instruction for all students, they must learn to recognize
students' views of classroom phenomena. The purpose of this study was
to identify the definitions of reading constructed by students in one
Significance of the Study
Unlike much research and practice in reading instruction which
focus on cognitive aspects of reading, the focus of this study is on
the social milieu in which the learning process occurs. Why study
children's perceptions of reading from this perspective? Researchers
who have examined classrooms as social environments have uncovered
multiple, differentiated forms of social organization within individual
classrooms. The classroom has been shown to be an extremely complex,
dynamic environment in which interacting students and teachers con-
struct and abandon contexts from moment to moment. The social-interaction
research perspective broadens the traditionally held notion of "per-
ceptions" by viewing them as interactionally constructed products of
classroom contexts. An examination of social contexts, then, can
reveal the perceptions children construct and utilize in the classroom.
McDermott (1977) has pointed out that educational researchers
"have virtually ignored the social context of reading activities"
(p. 154). By focusing on the many contexts in which classroom reading
occurs, the present study attempts to investigate "the work that teachers
and students do together to construct, maintain, and modify their
definitions and conceptions about reading" (Anang, 1982, p. 1). A
careful examination of the social milieu of teaching and learning will
provide insight into the perceptions children construct as well as the
processes by which perceptions are constructed.
The study may yield a number of contributions to both research and
practice in the area of reading. For researchers the study has
methodological significance in that it illustrates the use of a per-
spective and related methods not commonly used in reading research. It
is likely that in addition to yielding products and providing insight
into classroom processes, the study will serve to highlight variables
which can be examined in future research. Further, a study of the
social contexts of reading adds to the small but growing body of re-
search which examines reading and readers in natural settings. By
integrating studies of the external contexts of reading, such as this
one, with studies of the internal or cognitive contexts of reading,
researchers may establish a more fully developed perspective of the
processes of reading and learning to read.
The study will also be of value to practitioners. Detailed de-
scriptions of the studied classroom, the teacher, and the students
may be familiar to many teachers. Teachers may see themselves and
their students in the illustrations included throughout the report of
the study. The recognizability of many features of the studied class-
room contributes to the consciousness-raising value of the study. That
is, the study may help teachers become aware of the dynamic interplay
among classroom participants and the importance of monitoring children's
perceptions of classroom phenomena. Attention to children's percep-
tions can lead to improved instruction, as teachers who become sensitive
to children's ways of thinking are better prepared to provide suitable
learning experiences. Additionally, insight into children's perceptions
of reading can help teachers interpret children's reading behavior.
Weinstein (1983) wrote of the practical value of this kind of research,
It is important for teachers to come to know the world of
school from the perspective of students. Being aware of
students as active interpreters of classroom events forces
teachers to examine more closely the effects of their own
behavior on the recipients of these interventions. (p. 302)
Definition of Terms
While research into children's perceptions of reading has tradi-
tionally been guided by a cognitive-developmental view, some recent
investigators have begun to explore this area from a social-interaction
perspective. Those who have assumed a cognitive-developmental view have
focused on the internal, mental context of reading and reading instruc-
tion, while those who have assumed a social-interaction view have turned
their energies to the external, social contexts of reading. From the
social-interaction perspective "perceptions of reading" are more
accurately thought of as interactionally constructed "definitions of
reading." Below, the terms "cognitive-developmental perspective,"
"social-interaction perspective," "context," and "definitions of read-
ing" are clarified.
Guthrie and Kirsch's (1984) description of the traditional view of
literacy is helpful in characterizing cognitive-developmentally oriented
studies of children's perceptions of reading. First, these studies
view perceptions as cognitive structures that exist in children's minds.
Second, the studies assume that one such structure exists; that is, a
child has one kind of perception of reading. Related to this is the
assumption that a perception is either correct or incorrect. Further,
it is assumed that once individuals acquire the "correct" perception,
they utilize the perception in all contexts. In an effort to identify
universal patterns in the cognitive development of children's percep-
tions of reading, researchers have typically focused on identifying
perceptions under carefully controlled experimental conditions.
Unlike the cognitive perspective which focuses on "contexts in the
mind" (Cazden, 1982, p. 418), the social-interaction perspective focuses
on the contexts in which individuals act and the interactions among
individuals within those contexts. The theoretical orientation known
as symbolic interactionism provides a framework for this approach. A
basic principle of this theory is that in order to understand people's
behavior, one must discover the meanings or definitions they attribute
to the object, process, activity, or individual of interest. Defini-
tions are not viewed as inherent in objects or activities; rather, they
arise "out of the social interaction one has with others" (Blumer,
1969, p. 24). Blumer summarized the approach as follows:
Symbolic interactionism sees meaning as arising in
the process of interaction between people. The meaning of
a thing for a person grows out of the ways in which other
persons act toward the person with regard to the thing.
Their actions operate to define the thing for the person.
Thus, symbolic interactionism sees meanings as social
products, as creations that are formed in and through
the defining activities of people as they interact. (pp. 4-5)
From this perspective there is no single abstract, correct defini-
tion of reading but many definitions that are actively constructed by
individuals as they interact in various settings. "Human experience is
such," wrote Denzin (1978), "that the process of defining objects is
ever-changing, subject to redefinitions, relocations, and realignments"
(p. 7). These definitions determine the ways the individual behaves as
a "reader." In order to understand a child's reading behavior, then,
one must gain access to his or her definition of reading. Social-
interaction studies of children's perceptions of reading attempt to
discover interactionally constituted definitions through careful study
of the situations in which definitions are created. As Blumer (1969)
wrote of symbolic interactionism, this perspective "lodges its problems
in [the] natural world, conducts its studies in it, and derives its
interpretations from such naturalistic studies" (p. 47).
Context, as it is used in this study, refers to more than the
physical setting in which an event occurs. Students of social and
communicative environments have developed the notion of context to
refer to "the constellation of norms, mutual rights, and obligations
that shape social relationships, determine participants' perceptions
about what goes on, and influence learning" (Gumperz, 1981, p. 5).
Contexts are established as people interact with one another. Together
individuals define the situation and rules for appropriate participation.
In the classroom, contexts can change from moment to moment. Children
who engage in reading throughout the school day (Griffin, 1977) may
construct a number of definitions which serve to guide their behavior.
Definitions of reading refer to the outcomes of the active process
children engage in when assigning meaning to the things of their world.
From a social-interaction perspective individuals construct definitions
through their interactions in social contexts. Characteristics of the
context contribute to decisions about "what counts as reading" (Heap,
1980). According to Heap these characteristics include
who the speakers and hearers are; who they take each other
to be; how much they know about each other; how much they
know that the others know about them; their reasons for
interacting, for doing, whatever they are now doing together;
their beliefs and assumptions about what they are doing
together. (p. 283)
In the multiple and changing contexts of the classroom, children and
teachers engage in an ongoing process of defining reading. Awareness
of established definitions helps in interpreting and understanding
individual and group behavior.
Design of the Study
Having received approval from the University Committee for the
Protection of Human Subjects and the county school board, the researcher
established an observation schedule with a first-grade teacher who had
previously agreed to have the study conducted in her classroom. Obser-
vations began during the public school preplanning week in August and
continued until the December vacation. The researcher observed 150
hours of classroom activity representing all days of the week and times
during the school day. Observations were focused on children's speech
messages about reading, their reading-related behavior, and their use
of reading materials. Formal and informal interviews were also con-
ducted throughout the observation period. Those interviewed included
the teacher, the children in the low and high reading groups, and the
children's kindergarten teachers. In addition, children's cumulative
school records were examined.
Data analysis was conducted throughout the study as described by
Spradley (1980). Data were organized into categories or domains based
on similarities among recorded events. Taxonomies were then constructed
to represent children's definitions of reading by drawing data from
across domains. Taxonomies, then, contained data from several differ-
ent domains all of which served to indicate children's definitions of
Scope of the Study
This study was conducted in one first-grade classroom and focused
on the reading-related behavior of approximately half of the students
in that room. These students included the nine members of the low
reading group and the six members of the high reading group. Observa-
tions and interviews were restricted to the first four months of the
school year. Had the study been conducted in the same classroom from
January until May, different findings may have resulted due to striking
curricular and instructional changes which had been implemented.
Although the study can provide insight into teaching and learning
processes, specific findings about these children's definitions of
reading should not be generalized to other populations.
Review of the Literature
A review of the literature on children's perceptions of reading
provides necessary background for the questions raised in the present
study. The review is organized into studies based on the cognitive-
developmental perspective and those based on the social-interaction
perspective. Following the review, guiding questions for the present
study are clarified.
Studies of Children's Perceptions of Reading:
Downing (1979) noted that reading specialists were slow to recog-
nize the significance of Piaget's (1959) and Vygotsky's (1962) findings
about children's language perceptions. Piaget found that children of
beginning school age had little awareness of the functions of communi-
cation. In the area of written communication, Vygotsky found young
children to have only vague ideas about the usefulness of writing.
Reading researchers did not begin to examine children's perceptions
of reading until the late 1950s. The relevance of their findings for
a theory of how children learn to read was not realized for at least
Using interviews and paper-and-pencil tests, several researchers
in the 1950s and 1960s attempted to assess children's perceptions of
reading. In an early study, Edwards (1958) asked questions of second,
third, and fourth graders of normal to superior intelligence, assigned
to remedial reading classes, to identify their definitions of a "good
reader." The subjects shared the perception that good reading was a
matter of speed and fluency. Edwards concluded that "children could
form a concept of the reading process which is not the same as the one
held by the teacher" (p. 24). As a misconception about the nature of
good reading could lead to reading difficulties, Edwards recommended
that the teacher must "make certain that the child at no time loses
sight of the true purpose of reading, the getting of meaning" (p. 241).
Later, Denny and Weintraub (1963) reviewed some of the early studies
and concluded that "almost nothing is known of how the beginning reader
sees the reading act" (p. 363). The following review of unpublished
and unavailable studies by McConkie, Muskopf, and Edwards is based on
Denny and Weintraub's report.
McConkie (as cited in Denny & Weintraub, 1963) interviewed 81
kindergarten children to discover their perceptions of reading and
possible differences in the perceptions of boys and girls. Denny and
Weintraub summarized the findings as follows: "Great variability was
found in children's ability to define reading; however, almost all
children were able to verbalize some concept about the meaning of
reading" (p. 363).
Denny and Weintraub reported that Muskopf examined the relation-
ships between first graders' concept of reading and intelligence,
reading achievement, and the instructional approach used by their
teachers. After having administered a forced-choice, paper-and-pencil
test to first graders at the end of the school year, Muskopf concluded
that there was no significant correlation between a child's concept of
reading and his/her reading achievement. He acknowledged the question-
able validity of the instrument used to measure concepts of reading.
Finally, Denny and Weintraub described Edward's study of the rela-
tionship between fifth-grade children's concept of reading and their
reading achievement. Only a slight correlation between the two vari-
ables was found. As was the case in the Muskopf study, Edwards pointed
out the questionable validity of the self-constructed reading concept
Having reviewed the meager data available, Denny and Weintraub
concluded, "There is a crucial need for more information about the
beginning reader's concepts of the reading act and about his insights
into himself as a potential reader" (1963, p. 364). They outlined a
proposal for an interview study to be conducted with first graders at
the beginning and end of the school year.
Weintraub and Denny (1965) reported the findings from beginning of
the year interviews of first graders. The researchers grouped children's
responses to the question "What is reading?" into seven categories.
They found that more than 25% of the subjects "failed to verbalize an
intelligible idea of the reading act" (p. 327). The three major
response categories were (a) object-related responses (33%), such as
"Reading is when you read a book"; (b) vague responses or those in which
children said they did not know (27%); and (c) cognitive responses
(20%), such as "reading is how to read and how to learn things." The
researchers found only minor response differences between boys and
girls. Based on their findings, they suggested that teachers should
help children to think of reading as "a thinking, meaningful act"
Denny and Weintraub (1966) also asked beginning first graders the
following questions: (a) Do you want to learn to read? Why? (b) What
must you do to learn how to read in first grade? They summarized their
findings as follows: "A fourth of all those entering first-graders
could express no logical, meaningful purpose for learning to read and
a third of the children had no idea how it was to be accomplished"
(p. 447). The reasons children gave for wanting to learn to read in-
cluded (a) wanting to read for themselves and to others; (b) wanting to
achieve a goal, such as "become smart"; (c) identifying with someone
who was a reader; and (d) placing a value on reading, suchas, "It's fun."
In response to the second question, the majority of responses were
placed in an "I don't know"/vague response category. The others were
obedience-oriented ("Do what the teacher says"), other-directed
("Teacher will show us how"), or self-directed ("Read to myself").
The researchers did not draw conclusions from this study which was
designed to be exploratory and descriptive.
Reid's (1966) study of 12 five-year-old children in Scotland explored
the development of reading and writing concepts during the first year
of schooling. She interviewed subjects after two, five, and nine
months in school, each time asking a different set of questions. Her
finding that "reading is a mysterious activity, to which [children]
come with only the vaguest of expectancies" (p. 60) was replicated with
a group of children in England (Downing, 1970, 1971-1972). Believing
that the reproduction form of response required by Reid might be dif-
ficult for young children, Downing (1970) added two procedures to hisstudy
of 13 four- and five-year-old children. First, Downing interviewed the
subjects after two months of school. Following the interview he had
subjects respond to questions about concrete stimuli such as color
photographs. For instance, subjects were asked to sort photographs into
"reading" and "not reading" categories. In the final task, subjects
were presented with auditory stimuli and asked yes/no questions about
the stimuli. For the first set of stimuli, subjects were to say "yes"
if they heard a word and "no" if what they heard was not a word. For
the second set of stimuli, they repeated the procedure for the concept
"sound." Downing concluded that, indeed, young children have a vague
notion of the purpose of reading and of the activities involved in
reading. Concerning the methodology utilized, Downing concluded that
children demonstrated more advanced ability in the presence of concrete
objects than in the interview situation.
Downing (1969) was motivated to pursue investigations of children's
perceptions of reading by his conviction that "children's thoughts about
reading, their notions or conceptions of its purpose and nature, present
the most fundamental and significant problems for the teacher of
reading" (p. 217). Downing and his colleagues (Evanechko, Ollila,
Downing, & Braun, 1973) developed an instrument to measure reading
readiness and to determine the best group of subtests for predicting
end of first grade reading achievement. The battery was administered
to 97 first-grade children. The four tested areas included concept of
the reading task, perceptual ability, linguistic competence, and
cognitive functioning. It was found that performance on subtests in
all four areas predicted success in reading. The authors further con-
tended that for a readiness test to serve a diagnostic function, it
should have a range of subtests representing the four general areas
Blanton and Mason (1970) investigated the relationship between
knowledge about reading and later achievement. These researchers found
a significant relationship between 5 year olds' scores on the Individual
Reading Interest Survey and their scores on the Wide Range Achievement
Test. They suggested that teachers must "ascertain what information
and beliefs about reading are held by their students before blindly
plunging into reading readiness activities or reading instruction"
Following the development and testing of their first reading
readiness survey, Downing and his colleagues decided to revise and
extend the subtests in the original battery to create a new test that
would focus exclusively on children's conceptions of literacy. Ayers
and Downing (1982) reported on the development of the new instrument,
the Linguistic Awareness in Reading Readiness (LARR) Test (Downing,
Ayers, & Schaeffer, 1982). Reliability and validity were established
by administering the LARR Test to kindergarten children and following
up in first grade with a test of reading achievement. The LARR Test
was found to be a significant predictor of later reading achievement.
In other studies, Downing, Ollila, and Oliver (1975, 1977) investi-
gated the relationship between children's reading concepts and their
pre-school experiences. Results of studies with Canadian Indian
children (1975) and with children representing a range of socio-
economic levels (1977) led the researchers to conclude that "experience
at home is an important factor in learning the purposes of reading and
writing" (Downing, 1979, p. 12).
In a study of Native American Headstart children's language con-
cepts, Oliver (1975) reached a conclusion similar to that expressed by
Downing et al. Oliver found that home experiences seemed to be related
to children's reading and writing concepts. Based on task performance and
interview data from 78 three-, four-, and five-year-old children, Oliver sug-
gested that "experiences with books, learning activities, watching
television, and interacting with other children seem to have had more
effect on concept building than did age" (pp. 868-869).
During the same time period in which Downing conducted his earliest
studies, Johns (1970) was asking children the question, "What is
reading?" Over three years he asked the question of children in fourth,
fifth, and sixth grades. Their responses led Johns to suggest "that a
better understanding of reading should be acquired by elementary
children" (p. 657). He then posed the question, "Would it not be
beneficial, then, to tell children what reading is all about?" (p. 648).
In his next investigation of children's concepts of reading, Johns
(1972) explored the relationship between reading concepts and reading
achievement. Fifty-three fourth graders were interviewed individually,
and their responses to the question "What is reading?" were recorded.
Responses were then classified into one of the following five cate-
gories: (a) no response, "I don't know," or a vague response;
(b) classroom procedures, such as, "You read a story and do workbook
pages"; (c) word recognition, such as, "Saying words"; (d) meaning or
understanding, such as, "It's when you read a story and know what it's
about"; and (e) meaning and word recognition, such as, "You learn the
words and read the story and you're supposed to know what it means."
Reading achievement was determined by administering vocabulary and
comprehension subtests of the Gates-MacGinitie Reading Tests Survey D.
A significant positive correlation was found between the children's
concept of reading and their reading achievement. Johns concluded that
the results were encouraging and that children's concepts of reading
warranted further investigation. "It may be," wrote Johns, "that one
of the contributing factors to children's reading achievement is their
understanding of the reading process" (p. 57).
In an effort to compare the concepts of reading held by good and
poor readers, Johns (1974) asked 103 fourth and fifth graders, "What
is reading?" Responses were then sorted into the five categories out-
lined above. The comprehension subtest of the Gates-MacGinitie Survey D
was administered to identify good and poor readers. Good readers were
those children who scored a year or more above grade placement; poor
readers scored a year or more below grade placement. Johns found that
good readers gave significantly greater number of meaningful definitions
of reading (in categories three, four, and five) than poor readers. He
identified several questions needing careful reflection in the future,
among them, "What should a meaningful concept of reading include?" and
"How should children's concepts of reading be explored?" (p. 60).
In a subsequent study Johns and Ellis (1976) investigated the
views of reading held by 1,655 students in grades one through eight.
Children were individually interviewed and asked the following ques-
tions: (a) "What is reading?" (b) "What do you do when you read?" and
(c) "If someone didn't know how to read, what would you tell him/her
that he/she would need to learn?" (p. 119). Children's taped responses
were assigned to the same a priori categories listed above. Responses
for each question were analyzed for general trends and differences in
sex and grade. The researchers identified five major conclusions:
1. Many students have little or no understanding of
the reading process.
2. Older students have a somewhat better understanding
of the reading process than younger students.
3. There were few sex differences in the data. How-
ever, when differences existed it was revealed that boys
gave more vague or irrelevant responses than girls. Also,
girls appeared to be more aware of the fact that decoding
and meaning were essential for reading.
4. Most of the meaningful responses described reading
as a decoding process. It may be that teachers are over-
emphasizing decoding or "sounding out" strategies to the
exclusion of the role meaning plays in reading.
5. Many children have a very restricted view of
reading. They described reading as an activity occurring
in the classroom or school environment which utilized a
textbook. (pp. 125-126)
As a practical implication of their findings Johns and Ellis (1976)
recommended that teachers help students "grasp a worthwhile concept of
reading" (p. 126). For researchers the authors recommended investiga-
tions of the effects of teaching children a concept of reading on their
reading achievement. Further, they suggested that a variety of tech-
niques be utilized to explore children's concepts of reading. Their
examples included in-depth interviews and questionnaires.
Tovey's (1976) interview study of children's perceptions of read-
ing assumed a different focus from the studies which preceded it. Tovey
designed his questions to reflect several features of a psycholinguistic
approach to reading. He was specifically interested in children's
perceptions of reading as a silent process, as a process of deriving
meaning from written language, as a predictive process, and as a pro-
cess utilizing three cue systems. During 15- to 20-minute individual
interview sessions Tovey questioned 30 children, five each from grades
one through six. His findings were summarized as follows: (a) Children
perceived reading as an oral activity; (b) One-fifth indicated that
reading had something to do with meaning, while the largest percentage
spoke of reading in terms of decoding; (c) The majority of children
expressed the view that each letter and word must be processed to obtain
meaning; and (d) Most of the children perceived the graphophonic cue
system as the only strategy for decoding print. Tovey concluded that
children's responses reflected the way they had been taught to think
about reading. The findings imply "that teachers are using the 'word
recognition equals reading' model" (p. 540). He advocated that children
be encouraged to use psycholinguistic concepts and processes as they
learn to derive meaning from text.
In a study of preschool children's print awareness, Hiebert (1981)
used a different methodology from earlier studies and found different
results from those studies (e.g., Weintraub & Denny, 1965; Reid, 1966).
Hiebert examined two areas: (a) children's reading readiness skills,
such as visual discrimination; and (b) their knowledge of processes
involved in and purposes of print. Her objectives were to establish
developmental patterns of print-related concepts and skills and to
determine interrelationships among the concepts and skills. To tap
children's knowledge of print processes and purposes, Hiebert designed
tasks which utilized meaningful stimuli in concrete situations. For
example, in one task the investigator read orally from a book and then
asked the child to name the activity he/she had just observed. Unlike
earlier researchers, then, Hiebert presented her 3-, 4-, and 5-year-old
children with concrete reading-related tasks in meaningful contexts.
Hiebert found that "when exposed to print within meaningful con-
texts, children in the present study seemed quite aware of print and
its use" (1981, p. 254). While Hiebert's purpose was to identify
developmental patterns in print awareness, she was particularly sensi-
tive to the impact of socio-cultural factors on the developmental
process. In addition to the influence of meaningful contexts, Hiebert
pointed out the impact of time and place on opportunities to learn about
print. The popularity of Sesame Street and other children's television
programs, the proliferation of signs and labels in a child's environ-
ment, and the abundance of children's books may have contributed to the
greater awareness of print demonstrated by her subjects than subjects
in the 1950s and early 1960s. Pointing to her findings that readiness
skills and concepts appear to develop in an integrated fashion as
children accumulate experiences with print, Hiebert advocated that
"information used to structure reading experiences should come from
careful documentation of what children actually do in naturalistic and
school settings" (p. 257).
In an effort to understand why poverty children so often had
difficulties with beginning reading, Ferreiro and Teberosky (1982)
interviewed Argentinean first graders at the beginning, middle, and end
of the school year. The authors clarified their perspective by noting,
Obviously we are aware of the presence of factors external
to schooling that are involved in failures, but we believe
that there are also internal factors--directly related to
the external ones--that stem as much from the conception of
learning as from the objective purposes the school hopes to
achieve. (p. 90)
The authors' view was that children construct conceptualizations about
the nature of written language just as they construct understandings of
other phenomena of their world. This Piagetian perspective guided the
authors in identifying developmental stages in children's conceptions
of a variety of features of reading. Among the areas examined were
children's understanding of letters and punctuation marks and their
understanding of the relationship between drawing and print. Children's
responses to a variety of tasks and questions indicated that their con-
ceptions of the nature of reading represented a range of levels of cog-
nitive development. The authors pointed out that school instruction
tends to be based on adult definitions of reading concepts and processes.
As a result, only those children who have achieved more advanced levels
of conceptualization can benefit from traditional instruction. Other
children "will have more difficulty reconciling adult proposals with
their own hypotheses about written texts" (p. 89). Recognition of
the existence of differing conceptualizations requires that teachers
rethink some of their most taken-for-granted assumptions about teaching.
For example, before attempting to teach children a set of "easy" sight
words, a teacher must ask, "Easy for whom, easy from whose point of
view, from whose definition of easy?" (p. 51).
Saracho (1983) conducted a study of children's perceptions of
reading to identify factors that influence the reading performance of
bilingual bicultural children. This researcher's goal was to analyze
4- through 8-year-old Mexican American children's perceptions in terms
of their cognitive styles. She found that "young children in the study
were insensitive to or unaware of many important dimensions of read-
ing" and suggested, "Since knowledge about reading is necessary for
the acquisition of reading skills, educators may need to incorporate
programs to teach young children to realize that the main purpose of
reading is to abstract meaning from the written word" (p. 217).
Saracho's analysis of developmental trends in perceptions revealed that
children's perceptions of reading were "more closely related to their
experience with reading than to anything else" (p. 217), including
developmental shifts. She recommended that teachers use information
about children's cognitive styles as well as their perceptions of the
reading process to improve instruction.
A recent study investigated the effects of instruction on kinder-
garten children's perceptions of the nature and purpose of reading.
Mayfield (1983) used a code systems approach to provide children with
25 minutes of daily instruction for 20 days. This approach was de-
signed to help children explore code or symbol systems in their environ-
ment, to learn concepts and vocabulary related to code systems, and to
consider the uses of code systems. Pre- and post-treatment measures of
children's perceptions of reading were obtained through interview
questions and performance on several subtests of the Evanechko et al.
(1973) reading readiness test. Mayfield's findings indicated that code
system instruction led to improved, more adult-like perceptions of
reading. Regarding methods of identifying children's perceptions,
Mayfield concluded that interview questions alone were probably not
useful in obtaining accurate perceptions. Her observation that children's
responses to interview questions were frequently contradicted by their
performance on test items led her to point out that children might
understand a concept without being able to verbalize it. Mayfield
recommended that instruments designed to identify perceptions of read-
ing be improved and that code systems instruction be integrated into
The reviewed studies have contributed to our knowledge of children's
perceptions of reading. It is becoming increasingly clear that chil-
dren's concepts of reading are positively related to their reading
achievement (Blanton & Mason, 1970; Evanechko, Ollila, Downing, & Braun,
1973; Johns, 1972; Johns & Ellis, 1976). A number of the studies have
indicated that children's concepts of reading are strikingly different
from teachers' concepts (Edwards, 1958; Ferreiro & Teberosky, 1982;
Johns, 1972; Weintraub & Denny, 1965). Many children seem to perceive
of reading as a decoding process (Johns & Ellis, 1976; Tovey, 1976),
while their teachers believe that text comprehension is the goal of
reading. Home experiences with reading have been found to be related
to children's perceptions (Downing, Ollila, & Oliver, 1975, 1977;
Oliver, 1975; Saracho, 1983), and developmental stages in children's
perceptions of reading have been identified (Ferreiro & Teberosky, 1982).
Although some researchers claimed that young children have a vague notion
of the purposes of reading (Downing, 1970, 1971-1972; Reid, 1966),
others demonstrated that even preschool children were aware of print
uses when presented with tasks in meaningful contexts (Hiebert, 1981).
Despite these contributions there remain gaps in our understanding
of young readers and their views of reading. For example, we do not
know why children develop the perceptions of reading that they do. In
addition, we do not know whether children use one perception of reading
in all settings or whether, as suggested by Hiebert (1981), their per-
ceptions are related to the context at hand. We also do not know why
there seem to be differences in the perceptions of good and poor
readers. Furthermore, we know little about the role of children's
perceptions of reading in the ongoing life of the classroom. These
remaining questions are due in part to the methods which have been used
to investigate children's perceptions. Most of the reviewed studies
had several features in common. One, they all attempted to identify
children's understanding of reading at a particular moment in time
rather than over time. Two, they all relied heavily on interviewing
and paper-and-pencil tests. Three, they all removed subjects from
natural settings. In an effort to identify and to generalize children's
thoughts about reading, the contexts in which thinking occurred tended
to be neglected. When context was addressed, it was found to be a sig-
nificant factor in children's perceptions of reading (Hiebert, 1981).
A small group of recent studies has begun to address some of the
noted gaps in previous research findings. In the following section a
second and much smaller set of studies of children's perceptions of
reading is reviewed. These studies represent a perspective which focuses
on the contexts in which behavior takes place.
Studies of Children's Perceptions of Reading:
A growing body of naturalistic studies have explored reading from
a social-interaction perspective. For instance, researchers have
studied the operation of reading groups (e.g., McDermott, 1976), the
development of literacy in children (e.g., Heath, 1983; Taylor, 1983),
and the experiences children have with reading throughout the school
day (Griffin, 1977). However, very few researchers have attempted to
study children's perceptions of reading from this perspective. In the
following section these few studies are reviewed.
Roth's (1980, 1983) 10-month study of 10 children in two first-
grade classrooms focused on the "meanings" or "sense children come to
make of reading over time" (1983, p. 5). Participant observation
and interviewing were used to collect data before, during, and after
the first-grade year. Through her analysis of the teachers' and students'
behavior patterns, Roth discovered the shared definition of reading
developed in each classroom. In one room "reading is a teacher-determined
task which: 1) needs to be completed within a given time period and
2) entails remembering what is read in order to repeat it for the
teacher upon request" (1983, p. 10). In the second classroom "reading
is a teacher-determined task which is done in order to: 1) understand
the meaning of what is read and 2) share how it relates to personal
experience" (1983, p. 11). Although before-school interviews indicated
that children eagerly anticipated learning how to read in the first
grade, as the year progressed children began to see reading in terms
of required tasks, especially paper-and-pencil activities. According
to Roth (1983), "One of the most telling indications of the impact of
schooling is seen in the way the children take on their teachers'
definitions of reading" (p. 11).
The relationship between teachers' behavior and students' concep-
tions of reading comprehension was the focus of a study by Mosenthal
(1983). More specifically, Mosenthal wondered whether differences in
the ways teachers organized reading instruction influenced the ways
students came to understand appropriate classroom comprehension.
Two fourth-grade teachers representing distinct teaching ideologies
were identified. The first reflected an academic ideology. This
teacher's instruction was characterized by features such as close ad-
herance to the teacher's manual and frequent rejections of student
responses drawn from sources other than the specific text at hand. The
second teacher's instruction reflected a cognitive-developmental
ideology. This teacher tended to adapt and modify the teacher's manual
and was likely to accept answers based on students' prior knowledge or
on past or future text. To ascertain the students' definitions of
appropriate reading comprehension, story-related questions were con-
structed for each teacher's low, middle, and high reading groups. Once
each group finished reading a section of a story aloud, the teachers
asked a set of experimenter-constructed questions. The students'
responses were classified according to the sources used to answer them.
For instance, was the response based on prior knowledge, prior text, or
current text? When a response appeared to be based on prior knowledge,
the subject was interviewed to determine whether this was in fact the
case. Responses were also categorized according to whether they were
identical to text, inferrable from text, or an embellishment of the
text. Comparisons were then made between students' responses in the
Mosenthal found that significant differences existed between the
kinds of responses given by students in the two classrooms. Students'
responses reflected their definitions of reading comprehension, defini-
tions they shared with their teachers. Mosenthal summarized as follows:
The importance of this study is that it demonstrates the
need to consider more than text or reader variables in
describing how children acquire reading competence. .
the study demonstrates that teacher ideology, as reflected
in the organization of classroom social situation, is an
important variable influencing children's reading
acquisition. (p. 546)
In a study of junior high students, Bloome (1982) investigated
in-school and out-of-school contexts in which students engaged in
reading. His purpose was to discover the definitions of reading con-
structed by students as they interacted in various contexts. So as to
avoid imposing his own definitions on participants' behavior, Bloome
(1980) considered reading to be indicated by an activity meeting any of
the following criteria:
1. defined as reading by participants
2. eye gaze in the direction of print
3. an instructional lesson defined as a reading
4. the potential use by participants of printed
symbols to communicate
5. communication about something read or to be
read. (p. 6)
Bloome stressed that the criteria were "merely signals that a reading
activity may be occurring" (p. 6).
Bloome collected data over an 11-month period through the use of
ethnographic techniques such as participant observation, videotaping,
and ethnographic interviews. Although a complete description of find-
ings is forthcoming, some information is already available (Bloome,
1982). For example, a close analysis of students' postural configura-
tions revealed a continuum of reading contexts called the social-
isolated reading continuum. One end of the continuum, isolated reading,
was typical of reading done inside the classroom. Isolated reading was
characterized by one student interacting with one text. Bloome found
that "activities officially labeled READING seemed to be primarily
'isolated reading'" (p. 22). The other end of the continuum, social
reading, was characterized by several students interacting with a
single text and one another simultaneously. This kind of reading was
most often found outside of the classroom. Bloome suggested the possi-
bility that people may define reading as a social activity involving
"the structuring of relationships between people, the establishing of
norms for participation, etc." (p. 25). According to Bloome, "reading"
may be an even more complex phenomenon than investigators have thought.
He suggested that insight into the ways in which teachers and learners
define reading may have implications for instruction, evaluation, and
The three reviewed studies focused on identifying interactionally
constructed definitions of reading. While Roth and Mosenthal concen-
trated on elementary-aged children who adopted their teachers' defini-
tion of reading, Bloome studied junior high students as they defined
reading in multiple contexts. These studies have begun to fill in some
of the gaps left by the earlier studies. For example, they provided
insight into why children develop the perceptions of reading that they
do. In addition, the Bloome (1982) study provided evidence that chil-
dren constructed multiple definitions of reading as they interacted in
various contexts. Nevertheless, questions about children's definitions
of reading remain. It is still not clear whether children use one or
more than one definition of reading at a given stage in their develop-
ment. We also do not know how to account for the differences in the
perceptions of good and poor readers. Additionally, the three reviewed
studies suggest important questions about children's definitions of
reading in the classroom setting. Is there just one definition of
reading constructed in a classroom? If not, what kinds of definitions
are constructed? Do all classroom participants construct the same
definitions of reading? What consequences do children's definitions
have for their success or failure as readers?
The purpose of the present study is to address some of the un-
answered questions in this area of research. The following broad
questions served to guide the investigation:
1. What are the definitions of reading constructed by
members of the low and high reading groups in one first-grade
2. Are there patterns in children's definitions within
and across ability groups?
In the following chapters the methodology, findings, and implica-
tions of the study are discussed. In Chapter II, the methodology is
described. Chapters III and IV present the study's findings. Conclu-
sions and implications are discussed in Chapter V.
Qualitative research methods were used to investigate first-grade
children's perceptions of reading in this study. Bogdan and Biklen
(1982) outlined several features which characterize this kind of in-
quiry: (a) Research is conducted within the particular setting under
study; (b) The researcher is the main research instrument; (c) Data are
descriptive; (d) The focus is on ongoing processes rather than pro-
ducts; (e) Data are analyzed inductively; and (f) The researcher is
concerned with understanding the perspectives of the people under study.
The approach may also be called "naturalistic," as the objective was
"to illuminate social realities, human perceptions, and organizational
realities untainted by the intrusion of formal measurement procedures
or reordering the situation to fit the preconceived notions of the
investigation" (Owens, 1982, p. 7). A qualitative, naturalistic re-
search approach can yield "a literal description that figuratively
transports the readers into the situation with a sense of insight,
understanding, and illumination not only of the facts or the events in
the case, but also the texture, the quality, and the power of the con-
text as the participants in the situation experienced it" (Owens, 1982,
p. 8). Wilcox (1982) described qualitative research as "a naturalistic,
observational, descriptive, contextual, open-ended, and in-depth
approach to doing research" (p. 462). Why use such an approach to
study children's definitions of reading?
In this study the researcher assumed a social-interaction perspec-
tive grounded in Blumer's (1969) symbolic interactionist theory. Accord-
ing to this perspective, "objects are social products in that they are
formed and transformed by the defining process that takes place in social
interaction" (Blumer, 1969, p. 69). Reading, then, is not an abstract
construct with inherent characteristics, but a product of the inter-
actions of individuals in various contexts. In order to understand
children's definitions of reading, the researcher must closely examine
the social interactions within which definitions are constructed.
According to Denzin (1978), "Interactionists regard human interaction
as their basic source of data" (p. 7). A social-interactional investi-
gation of children's perceptions of reading must focus on the contexts
in which children have opportunities to work out the meanings of read-
ing. A qualitative research approach provides a methodology com-
patible with the social-interaction perspective and suitable for the
investigation of interactionally constituted definitions of reading.
According to Blumer (1969), "No theorizing, however ingenious, and
no observance of scientific protocol, however meticulous, are substi-
tutes for developing a familiarity with what is actually going on in
the sphere of life under study" (p. 39). Blumer advocated the disci-
plined use of exploration and inspection in the examination of the
social world. He conceived of the naturalistic method as the
methodological perspective of symbolic interactionism. The following
quote summarizes Blumer's research approach:
Exploration and inspection, representing respectively
depiction and analysis, constitute the necessary procedure
in direct examination of the empirical social world. They
comprise what is sometimes spoken of as "naturalistic"
investigation--investigation that is directed to a given
empirical world in its natural, ongoing character instead
of to a simulation of such a world, or to an abstraction
from it or to a substitute for the world in the form
of a present image of it. Naturalistic inquiry, embracing
the dual procedures of exploration and inspection, is
clearly necessary in the scientific study of human group
life. (pp. 46-47)
In this study, interactions within a first-grade classroom were
explored and inspected through the use of qualitative, naturalistic
methods. These methods were borrowed from the disciplines of anthro-
pology and sociology and could be characterized as ethnographic. In
this chapter the researcher's methods and procedures are described.
Included are discussions of the setting, the research model and each
of its components, the issue of validity, and researcher qualifications
and biases. The first topic, the setting, is described in three sec-
tions: the selection of the site, gaining entry to the site, and a
description of the site.
Selection of Research Site
The study was conducted in one first-grade classroom. The selec-
tion of the classroom was guided by several criteria established on the
basis of project objectives. The criteria for classroom selection were
as follows: (a) a first-grade classroom, to ensure that all children
would be engaged in formal reading instruction from the beginning of
the school year; (b) a classroom in which children had opportunities to
interact and read throughout the school day; and (c) a teacher who was
comfortable having an observer in the classroom. An additional con-
sideration, though not a requirement, was that the teacher and researcher
already be acquainted. According to Lofland (1971), access to desired
information is more easily gained when the researcher uses "pre-existing
relations of trust as a route into the setting, rather than 'going in
cold'" (p. 95). It was believed that a harmonious researcher-teacher
relationship would be more easily established if a friendly rapport
already existed between the two. The researcher was acquainted with
the first-grade teachers in five public schools. The selected teacher,
Mrs. Saunders,1 met all criteria and was one with whom the researcher had
a friendly rapport.
Gaining Entry to the Site
Once the teacher was selected, the researcher visited her in her
classroom to make an appointment for an initial meeting. The meeting
took place in the spring. The researcher's goals for this meeting were
to explain project purposes and plans and to ask Mrs. Saunders' per-
mission to conduct the study in her classroom. A brief outline of the
proposed project had been prepared and was shared with the teacher (see
Appendix A). By discussing the study in general terms, the researcher
attempted to avoid influencing project outcomes. The lengthy data
collection period also minimized the possibility that this initial
The teacher's name and all students' names are pseudonyms.
discussion would have an impact on study findings. In addition to
explaining guiding questions and research methods, the researcher
explained her reasons for choosing the classroom. During the initial
meeting the researcher also described what her role would be in the
class. She explained that although she would be spending much of her
time writing and asking questions, she would make time to talk with
Mrs. Saunders to answer her questions and discuss concerns about the
study. The researcher also assured Mrs. Saunders that she would have
access to the study's findings. The teacher listened and responded
enthusiastically during the meeting. She expressed great interest in
the researcher's questions and agreed that they were important to in-
vestigate. She said she was "honored" that her classroom had been
selected, and she would be happy to be involved in the study.
On the same morning the researcher met with the school principal to
explain the proposed study and to get his informal permission to con-
duct the study at his school. A copy of the project outline was shared
with the principal. He agreed to the plan and reminded the researcher
that the project had to be approved by the district office before it
could be implemented.
Two sets of paperwork were required for official approval of the
project. In order to gain entry to the classroom, an application to
conduct research in the district public schools was submitted to the
school district office. A description of the proposed project and a
parental consent letter were submitted to the University's Committee
for the Protection of Human Subjects. By the end of June the project
was approved by the school district's director of research and by the
University Committee. Written parental consent was obtained from all
parents in September. The researcher contacted the teacher to inform
her of the project's approval and to make plans for meeting during the
public schools' preplanning week in August.
The purposes of the preplanning week meeting were to remind the
teacher of the objectives and methods of the study and to interview her
concerning her decisions and plans for reading instruction. During this
meeting the researcher established that she would observe in the class-
room for 10 hours each week until the December holiday. Although the
teacher claimed that she did not need to have an observation schedule
because she was accustomed to being observed, the researcher promised
to tell her ahead of time when she would be observing. The researcher
asked the teacher's advice concerning the best way to obtain parents'
signatures on consent forms. The teacher suggested that the researcher
attend and speak at the first parents' meeting in early September at
which time she could introduce herself and explain her project. Finally,
it was agreed during this meeting that the researcher would come to
observe the first day of school.
Description of the Site
The study was conducted in a public elementary school in a large
southeastern city. The 30-year-old school was in a neighborhood that
had been characterized until recently as middle class. In recent years
the growing number of rental units and the increasingly transient popu-
lation have caused some city residents to consider the community to be
lower-middle class. The 600-member student body was 70% white and 30%
black and represented families in middle to lower socio-economic groups.
The majority of students were bused to and from school.
In the school there were four classrooms for each grade, kinder-
garten through fifth. Although there was not a gymnasium, there was an
art room, music room, media center, auditorium, and a cafeteria. The
following full-time support teachers were on the faculty: curriculum
specialist, varying exceptionalities teacher, gifted teacher, guidance
counselor, music teacher, media specialist, and physical education
teacher. A one-half time art teacher and a four-fifths time speech
teacher were also available.
The studied classroom filled half of a large space which also
housed a kindergarten. A playhouse and a bookshelf formed the partial
barrier between the two classrooms. Children and adults frequently
passed from one classroom to the next to borrow materials, to send
messages, to use the bathroom, and to teach or join a lesson. As shown
in Figure 1, the classroom was typical of many primary rooms. A chalk-
board lined one wall, while individual "cubbies" lined another, and
jalousie windows covered the third. Children sat at individual desks,
the organization of which was shifted by the teacher several times
during the first half of the school year. Her preferred plan consisted
of three groups of desks, with the desks in each group pushed together
forming a single rectangular surface. The seating arrangement was
determined by friendships and work habits rather than by ability groups.
Although children spent much of the school day at their desks,
they worked and played at other locations, too. Reading groups met
daily at a semi-circular reading table. This table was in front of the
chalkboard and also served as the teacher's desk. Sometimes children
were permitted to work outside at a picnic table. A quiet area in the
back corner of the room was used as a work place and also as a play
area, particularly by girls playing school. A listening center con-
sisting of a table with an audio tape player and individual earphones
was often used by children who listened to a tape-recorded story while
reading along in the book. On a long table by the listening center
there were sometimes art materials which children could use after finish-
ing their work. Often children used the floor to play games and put
puzzles together. The playhouse was a popular choice for some of the
girls. During free play periods, many children stayed at their desks
or gathered around one child's desk to draw, read, and play games.
In the classroom, children's artwork was displayed more often than
their academic work. Seasonal and holiday art projects resulted in
decorations which hung on windows, bulletin boards, and the playhouse. On
a section of one wall the teacher had painted a math facts rocket.
Different levels of the rocket represented different levels of addition
and subtraction facts. When a child could say all the facts for a par-
ticular level, his/her nametag was raised to that level of the rocket.
The alphabet written in the county-adopted handwriting style was posted
above the chalkboard. Commercially produced posters representing the
short vowel sounds were hung on the chalkboard and later on the play-
To acquaint the reader with the operation of the classroom, a
typical day is summarized. The teacher arrived at school at 7:30 A.M.
From this time until the children came in at 9:00 A.M. she planned,
talked with the teacher next door, and took care of other school respon-
sibilities. Attendance and lunch count were taken after the children
came in. The "line leaders" for the day took the attendance and lunch
information to the office and cafeteria. During this time the rest of
the children read a book, drew, or talked quietly with one another. Then,
the teacher introduced the children to their morning work. Typically
they had five tasks or "smilies" required of them. Some smilies required
instruction while others were assigned as practice exercises. Throughout
the morning the teacher met with each of the four reading groups. While
a reading group met, the other students worked on their smilies. When
children had completed all their smilies, they turned in folders con-
taining all assignments. Once work was turned in, children could choose
an activity such as reading, drawing, a special art project, a puzzle,
or any number of educational games. During the morning the class left
at least once for half an hour to go to a special class such as physical
education, music, or art. Twenty minutes of outdoor freeplay occurred
before 11 A.M. During this time the teacher checked the work in
children's folders. From 11:00-11:30 the teacher taught a math lesson.
After math the class went to lunch.
Returning from lunch at 12:10, the next 45 minutes were used by
the children to correct errors in their morning work and by the teacher
to meet with individual children who needed extra help. The teacher
announced daily the names of the children who were "superstars," that
is, those children who completed all their smilies without errors. The
remaining two hours of the school day were filled with activities such
as sharing time, a teacher-read story, a whole-class phonics lesson,
a special class, outdoor freeplay, and a science or social studies
activity. The children were dismissed at 3 P.M., and the teacher left
around 5:00 P.M.
The teacher, Mrs. Saunders, was a white woman who had taught for
about 15 years in two states. She had a master's degree in early child-
hood education and, during the course of the study, she applied and was
accepted to a specialist program, also in early childhood. She had
taught in private and public schools, in primary grades and in pre-
kindergarten programs. Having taught in a university laboratory school
and a federally funded model program, she was accustomed to working with
a variety of adult observers. Mrs. Saunders had taught for nine years
in this county and for six years at this school.
In addition to teaching, Mrs. Saunders was involved in professional
organizations. She was an active member of the Association for Child-
hood Education International, helped plan for the organization's con-
ference, and presented a paper at the meeting. Mrs. Saunders was also
actively involved in an education honorary organization. Due to her
reputation as an effective teacher, she was often invited to speak at
meetings and conferences.
Mrs. Saunders was respected and admired on a personal as well as
a professional level. During the study it was clear that she had warm
relationships with her students and with a number of the other teachers,
and she was very popular with student teachers. Her sense of humor and
constant optimism added to her popularity with students and peers.
The relationship which developed between Mrs. Saunders and the
researcher was trusting and harmonious. Researcher and teacher
conversations covered a wide range of personal and professional topics,
some related to project goals, many unrelated. Mrs. Saunders talked
openly about her family, difficult periods in her life, personal and
professional failures and frustrations, fears and concerns about stu-
dents, and school gossip. She often asked the researcher if she would
come to lunch and once invited her to a colleague's birthday celebra-
tion. Her attitude toward the researcher was conveyed in the following
remarks she made after the fieldwork period had ended. The researcher
had an appointment with the teacher to conduct an interview and began
by apologizing for taking up so much of the teacher's time:
Mrs. Saunders: Oh no, you're not! I miss you so much!
I love to have you come. You're so interested. You
really care about the kids and what they're doing. I
loved having you here because there was someone to ap-
preciate what I was doing.
Mrs. Saunders seemed genuinely to enjoy having the researcher in her
classroom. Not only was she comfortable being observed, but she ap-
preciated the attention and the opportunity to talk about herself, her
teaching, and her students.
There were 27 students in Mrs. Saunders' classroom. Of the 17
girls and 10 boys there were 9 black and 18 white children. Eight chil-
dren received free lunch and one paid a reduced price. All children had
attended kindergarten. The children in the top and the bottom reading
groups were the focus of the study. The groups were determined by
children's performance on the reading subtests of the Metropolitan
Readiness Test, the Metropolitan Achievement Test, and by their scores
on criterion-referenced tests of the county-adopted basal reader series.
The six white girls in the top group began the year in the first of the
two second-grade books. In the bottom reading group there were nine
children--two white boys, one white girl, three black boys, and three
black girls. This group began the year reading in the first pre-primer.
Although the two groups used different reading books and workbooks,
they used some of the same reading-related materials. These included
a phonics workbook and worksheets based on weekly "basic reading
Research Methods and Procedures
Children's interactionally constructed definitions of reading were
explored and inspected through a qualitative and naturalistic research
approach. Spradley (1979, 1980), an anthropologist, has organized and
synthesized traditional ethnographic methods into a systematic set of
procedures called the Developmental Research Sequence. He referred to
the research sequence as a methodology designed for the investigation
of meaning. Spradley's research model was adapted for use in this
study, the goal of which was to discover the definitions or "meanings"
children made of reading (Roth, 1980, 1983) in one first-grade class-
The ethnographic research model is cyclic rather than linear in
nature. That is, unlike the traditional, experimental researcher who
identifies hypotheses, collects data to test the hypotheses, analyzes
the data, and draws conclusions, the qualitative researcher utilizing
this model engages in a cyclical process of questioning, collecting data,
recording data, and analyzing data. Throughout the course of the study,
the sequence of questioning, collecting, recording, and analyzing was
repeated. Questions served to direct observations and from observations
emerged questions to provide further direction. Data analysis was not
the culmination of the research act, but an integral part of the re-
Asking Ethnographic Questions
Questioning is a critical element of the research cycle because
the questions the researcher asks direct data collection and lead him/her
closer to the perspectives of the people being studied. Not only are
questions posed prior to the study, but new questions are formulated
throughout the research period. A characteristic of this research
approach is that it begins without precise hypotheses which may "close
off prematurely the process of discovery of that which is significant
in the setting" (Wilcox, 1982, p. 459). While attempting to transcend
the influence of "preconceived ideas" which may bias the outcome of the
study, the researcher formulates "foreshadowed problems" to direct his
investigation. Malinowski (1922) distinguished between preconceived
ideas and foreshadowed problems as follows: "Preconceived ideas are
pernicious in any scientific work, but foreshadowed problems are the
main endowment of a scientific thinker, and these problems are first
revealed to the observer by his theoretical studies" (p. 9). In this
study two broad questions were posed to provide a general framework for
1. What are the definitions of reading constructed by
members of the low and high reading groups in one first-grade
2. Are there patterns in children's definitions within
and across ability groups?
While foreshadowed problems served to guide the overall direction
of the study, other kinds of questions were asked during the course of
the research. These questions included descriptive questions, struc-
tural questions, and contrast questions. Each kind of question led to
a different kind of observation and was associated with a different level
of data analysis. The three types of questions are defined as follows:
1. Descriptive questions were asked during the earliest observa-
tions when the researcher had little knowledge of the life of the class-
room. These general questions included "What is the daily schedule?"
"Who are the people in this classroom?" "How is the physical space
utilized?" and "When and where do children read?" These questions led
to descriptive observations which enabled the researcher to develop an
initial description of the unfamiliar setting.
2. Structural questions were asked following initial data analysis.
The purpose of asking structural questions was to add depth to identi-
fied categories of behaviors, objects, places, and people in the class-
room. For instance, it was found that the teacher asked many questions
of students throughout the school day. The structural question posed
by the researcher was "What are all the kinds of questions the teacher
asks?" The researcher also observed that children read in various
places in the classroom. The resulting structural question was "What
are all the places in which children read?" These and other structural
questions were asked repeatedly, the goal being to discover as many
answers as possible. In searching for answers the researcher made
focused observations. This type of observation enabled the researcher
to narrow the scope of the research and to discover the larger and
smaller categories existing in the classroom.
3. Contrast questions were posed following further data analysis.
After focused observations had filled in elements in the categories of
interest, such as the kinds of questions the teacher asked, contrast
questions were asked to identify differences among the elements. For
example, among the kinds of questions teachers asked were quiz questions
("What sound does that 'e' make?") and comprehension-check questions
("Why would there be a safe in a bank?"). To ensure that the two kinds
of questions were distinct elements in the category of teacher questions,
the researcher asked the contrast question, "How are these kinds of
questions different?" The question led to selective observations in
which the researcher searched the fieldnotes and conducted additional
field observations looking for differences between the two kinds of
Spradley (1980) suggested that the three types of questions and
observations be thought of as a funnel. The descriptive questions and
observations are "the broad rim of the funnel" (p. 128); structural
questions and focused observations narrow the scope of the study and
are represented by the narrower part of the funnel; contrast questions
and selective observations are represented by the small, narrow opening
at the bottom of the funnel. Ethnographic questions and their related
kind of observations are "the basic unit of all ethnographic inquiry"
(Spradley, 1980, p. 73). The questions asked determine the type of
observations made; questions and observations shape the course of data
collection. What kind of data were collected in this study? What
methods did the researcher use to collect data? These questions are
addressed in the following section.
Collecting Ethnographic Data
The researcher's objective was to discover and describe the defini-
tions of reading constructed by first-grade children. Since definitions
were not as easily observed as concrete phenomena such as teacher and
student questions and the use of reading materials, the researcher
utilized other kinds of data as indicators of children's definitions
(Barton & Lazarsfeld, 1969; Becker, 1970). According to Spradley (1980),
people everywhere make use of three types of information to make in-
ferences about what others know. He summarized, "We observe what people
do (cultural behavior); we observe things people make and use such as
clothes and tools (cultural artifacts); and we listen to what people
say (speech messages)" (p. 10). In this study the researcher used a
similar method of collecting evidence and making inferences. Barton and
Lazarsfeld (1969) stated the logic behind this practice: "The under-
lying assumption .. is that a phenomenon which cannot be directly
observed will nevertheless leave traces which, properly interpreted,
permit the phenomenon to be identified and studied" (p. 170).
In this study the researcher used children's speech messages about
reading, their reading-related behavior, and their use of reading
materials as indicators of "the less easily observed phenomena," their
definitions of reading (Becker, 1970, p. 28). Ethnographic data, then,
included what Spradley (1980) called speech messages, cultural behavior,
and cultural artifacts. How were these data collected?
Three main methods were used to collect data. According to Wolcott
(1976), the use of a variety of modes of gathering information may be
seen as "a critical underlying aspect of ethnography" (p. 35). Pelto
and Pelto (1978) suggested that a characteristic of ethnographic research
is its multi-instrument approach. Furthermore, these authors asserted
that "examining cultural behavior with a variety of different approaches
greatly enhances the credibility of research results" (p. 121). Denzin
(1978) also advocated the use of multiple methods of observation, or
triangulation. "Triangulation," he contended, "is a plan of action
that will raise sociologists above the personalistic biases that stem
from single methodologies. By combining methods observers can
partially overcome the deficiencies that flow from one investigator or
one method" (p. 294). The three methods of data collection used in the
study were participant observation, informant interviewing, and unob-
trusive measures. First these methods will be described. Then, prob-
lems inherent in the methods and the manner in which the problems were
managed will be discussed.
According to Schwartz and Jacobs (1979), participant observation
is the principal tool of the qualitative, naturalistic method. This
field method requires that the researcher
directly [participate] in the sense that he has durable
social relations in [the social system under investigation].
He may or may not play an active part in events, or he may
interview participants in events which may be considered
part of the process of observation. (Zelditch, 1969, p. 9)
Wolcott (1976) pointed out that school settings offer few formal roles
for the researcher to assume and hence participate in the social scene,
as traditional ethnographers have done. However, the role claimed by
the participant observer, or the role which subjects assign him/her,
has consequences for what he/she will be able to learn. Wrote
Schwartz and Jacobs (1979),
Who you are and where you are within such a world have a
role in creating that world and in fashioning the colored
glasses through which you see it and it sees you. .
the initial social role (and/or status) adopted by the
participant-observer usually remains fixed throughout his
study. It will define for him and others the way in which
he is part of the social world which he is studying.
(pp. 50, 52)
In this study the researcher was a "known observer" (Schwartz &
Jacobs, 1979, p. 55). Mrs. Saunders was aware of the researcher's pur-
poses, and the students were told that the researcher wanted to find
out what children do in first grade. The researcher's identity was
described to the students as "the one who writes and asks questions."
It was believed that this role would allow the researcher to "best study
those aspects of society in which [she] is interested" (Gold, 1969,
p. 38). As Strauss, Schatzman, Bucher, Ehrlich, and Sabshin (1969)
reported, it was necessary to restate the researcher's identity several
times during the study to remind students how to interact with the re-
searcher. When students began to ask the researcher for help on assign-
ments, for example, the researcher or the teacher announced that the
students would have to ask someone else because the researcher was there
to write and ask questions.
The type of researcher participation with the people and in the
activities observed was characterized as passive (Spradley, 1980). Re-
garding passive participation Spradley wrote, "The ethnographer engaged
in passive participation is present at the scene of action but does not
participate or interact with other people to any great extent" (p. 59).
During the first month of the project, the researcher rarely interacted
with students. Typically she stationed herself on the outskirts of the
classroom and recorded observed activities. She avoided eye contact
with students and ignored those who attempted to get her attention.
However, as the study progressed, the researcher began to ask questions
of students to gain greater insight into their behaviors, speech mes-
sages, and use of artifacts. During this period of increased inter-
action the researcher alternated between remaining at a fixed location
and moving around the classroom. Throughout the study the researcher
interacted with the teacher. Strategies utilized in student and
teacher interviewing will be discussed in a subsequent section.
Although the earliest observations were directed toward a general
description of the classroom, the majority of observations were oriented
toward reading-related activity. There were a number of occasions
during the school day when children engaged in reading or talking about
reading. For instance, some children chose books as soon as they
entered the classroom in the morning. Reading usually took place during
the introduction and explanation of morning work. While the children
worked on morning assignments, more reading took place. Reading occurred
during reading groups and during free time. There was also reading
and reading-related talk during story time in the afternoon. The re-
searcher directed her observations to occasions such as these when
children were involved with reading. In addition, the researcher ob-
served during seven of the children's Tuesday afternoon library visits.
Observations in the classroom and in the library focused on the activi-
ties of the 15 children who were in the top and bottom reading groups.
The researcher observed 150 hours of classroom activity over a
four-month period in the fall of 1983. Observations were approximately
evenly distributed across all days of the week and times of the day.
A total of 48 observations were made on the following days: 10 Mondays,
10 Tuesday, 10 Wednesdays, 11 Thursdays, 7 Fridays. As the researcher
completed an observation, she told the teacher when she would return.
This manner of scheduling was requested by the teacher.
Two types of interviewing were utilized in this study: formal
interviewing and informal interviewing (Spradley, 1980). Formal inter-
views were those which occurred as a result of a request by the re-
searcher. In these cases the researcher had particular questions in
mind and asked the informants to schedule a time when they could meet
with her. Formal interviews were conducted with the teacher, the
teacher in the adjoining classroom, the kindergarten teachers of the
studied children, and the children. These interviews took the form of
"guided conversations" as described by Lofland (1971). Lofland sum-
marized the nature of guided conversations as follows:
the aim [is] to provide for oneself a list of things
to be sure to ask about when talking to the person inter-
viewed. It is because of this aim that this type of device
is called an interview guide rather than an interview schedule
or questionnaire. One wants the interviewee to speak freely
and in his own terms about a set of concerns you bring to the
interaction, plus whatever else the interviewee might
introduce. (p. 84)
Although core questions provided a framework for formal interviews, the
conversational nature of these interviews often led to additional ques-
tions and to unanticipated, volunteered information from interviewees.
For example, the following core questions were identified to guide
interviews with students in the top and bottom reading groups:
1. Do you see people reading now? How do you know they're
2. Who is the best reader in the class? How do you
3. What reading do you do when you're at school? At
home? Is there someone else at home who reads? Who? What
kind of reading does this person do?
4. Are you a good reader? How do you know?
5. Why do people read?
During one interview, a girl volunteered, "My mamma teached me [to read]
when I was five years old." This statement prompted the researcher to
ask, "Tell me what she taught you." This kind of conversational give-
and-take resulting in new questions and unexpected data was typical of
interviewing in this study. Core questions for formal interviews with
Mrs. Saunders, the teacher in the adjoining room, and the kindergarten
teachers are in Appendices B, C, D, E, and F.
Informal interviews occurred on those occasions when the researcher
asked questions of the children and teacher during the course of par-
ticipant observation. The questions were typically suggested by an
observed event rather than determined in advance. For example, when
children worked on seatwork assignments, the researcher asked such
questions as "What are you doing?" "How are you supposed to do it?" and
"Why did the teacher have you do this page?" The majority of informal
teacher interviewing took place during periods when the children were
not in the room. The researcher often asked the teacher to comment on
a child's behavior. For instance, the researcher frequently asked
questions such as, "What do you think was going on with Tommy during
reading group?" or "Why do you think Susie had so much trouble with the
vocabulary worksheet?" So as to avoid disrupting classroom activity,
the researcher refrained from questioning the teacher during instruc-
tional time unless the teacher initiated the interaction. Often when
the teacher initiated an interaction she did so to fill the researcher
in on an event the researcher had missed. The teacher was a valuable
informant in that she frequently volunteered information about children's
activities that the researcher had not been present to observe.
Webb, Campbell, Schwartz, and Sechrest (1966) defined unobtrusive
measures as "measures that do not require the cooperation of a [subject]
and that do not themselves contaminate the [data]" (p. 2). Denzin (1978)
explained that "the use of unobtrusive measures represents an awareness
on the part of sociologists that their presence as observers is foreign,
and therefore in some sense reactive" (p. 257). In other words, as
Schwartz and Jacobs (1979) explained, the investigator's presence and
activities affect the social process being studied. Unobtrusive measures
minimize the possibility that the observer's presence "may change the
very world being examined" (Schwartz & Jacobs, 1979, p. 75). The spe-
cific measures used included physical trace analysis and document
According to Denzin (1978), "physical trace analysis is most appro-
priately viewed as a strategy for recording the incidence, frequency,
and distribution of social acts toward certain social objects through
time and in various situations" (p. 260). In this study the researcher
noted both the children's and the teacher's use of reading-related
materials. She recorded children's choices of trade books during free
time and their subsequent use of chosen books. Children's use of assigned
materials was also recorded, and completed assignments were often ex-
amined when the children left the room. Use of other materials such as
word cards and lists, individual 10" x 12" chalkboards, drawing paper,
encyclopedias, and library books was also recorded. The teacher's
choice and use of instructional materials were noted as were the teacher-
student interactions that took place regarding reading-related materials.
The analyzed documents were the children's cumulative school re-
cords. Individual folders contained the following information:
physician's report; personal history, including developmental milestones,
social development, and interests; family data, such as address, phone
number, and parents' occupations; scores on the Metropolitan Readiness
Test or the Metropolitan Achievement Test, primer level; kindergarten
report cards, including the instructional strategy to which the child
was assigned on the basis of the state's Primary Education Program; and
results of screening for speech and gifted programs. It was felt that
these data would be helpful in supplementing the researcher's observations.
Although unobtrusive measures do not present the same problems as
participant observation and interviewing, they also do not permit much
insight into the perspectives of the people being studied. The researcher's
goal was to overcome the shortcomings of participant observation and
interviewing by triangulating the three methods. The combining of
methods is not only characteristic of ethnographic research but is
recommended by many qualitative researchers (Becker & Geer, 1969; Denzin,
1978; McCall & Simmons, 1969; Pelto & Pelto, 1978; Schwartz & Jacobs,
1979; Webb et al., 1966). What are the problems with participant ob-
servation and interviewing, and how were these problems managed? These
questions are addressed below.
There are problems inherent in participant observation which must
be addressed if the researcher is to have confidence in the quality of
the data collected. McCall and Simmons (1969) summarized three cate-
gories of threats to data collection as follows: "1) reactive effects
of the observer's presence or activities on the phenomena being ob-
served; 2) distorting effects of selective perception and interpretation
on the observer's part; and 3) limitations on the observer's ability to
witness all relevant aspects of the phenomena in question" (p. 78).
These problems may distort collected data such that the researcher's
written records do not represent the naturally occurring events of the
classroom. In this study procedures were incorporated to minimize the
potentially damaging effects of the problems of participant observa-
The first problem, observer effects on the behavior of teacher and
children, confounds almost all research (Bogdan & Biklen, 1982). The
researcher took several steps to minimize the problem. First, one of
the criteria used in selecting a classroom to be studied was that the
teacher be accustomed to having observers in the room. Not only was the
studied teacher comfortable with other adults in her room, but the
children were constantly exposed to other adults including university
students, volunteering parents, an aide, and observers from other
schools. The researcher's presence, then, was not unusual or intimi-
dating to either teacher or children. Another feature of the study
which helped the researcher determine the effects of her presence on
classroom activity was the length of time she spent in the classroom.
Observations began on August 17th during preplanning week and continued
through December 16th. This lengthy observation period enabled the re-
searcher to become a taken-for-granted part of the classroom. By re-
maining on the outskirts of classroom activity and avoiding eye contact
with children during the first weeks of the study, the researcher
quickly blended into the background. In addition, by reminding children
of the researcher's role--the one who writes and asks questions--the
teacher and the researcher made it clear to the children that the re-
searcher was not another teacher. One indication of the children's
lack of attention to her was their tendency to misbehave in her presence.
For example, the researcher sat in on a reading group conducted by a
student teacher, Ms. Clark. The following excerpt demonstrates that
the students were not inhibited by the researcher, nor did they per-
ceive her as an authority figure in the classroom. Had Mrs. Saunders
been observing the lesson, the students would not have misbehaved as
Ms. Clark: Okay, so you're going to read the words and the
sentences and put the words in the right sentence. Can
anyone do the first one?
Ellen: Can I get something to drink, Ms. Clark, please?
Ms. Clark: When we're finished.
Ellen: No, now. I'm thirsty.
Ms. Clark asks someone to do the next one.
Ms. Clark reads the sentence: "The food good." So
what is it?
Ms. Clark, to Ellen, who leans back in the chair and looks
around the room: Ellen, what will you write in number two?
Robin: Stupid. We write "stupid page."
Ms. Clark: Ellen, the longer you take, the longer it's going
to be before you can go to the bathroom.
Ellen: It's a drink I want!
Ms. Clark: Well, same thing.
Ellen: I have to go blow my nose. (she stands)
Ms. Clark: Ellen, here's a kleenex, right here. (Ellen,
Robin, and Tracey all take kleenex and start blowing their
Another advantage of the length of time spent observing in the
classroom was that the researcher could use her knowledge of the setting
and of children's typical behavior patterns to judge the naturalness of
observed events. Although the great majority of children's behavior
seemed to be unaffected by the researcher's presence, some episodes
reflected observer effects. For instance, children occasionally "showed
off" for the researcher, as the following brief excerpt illustrates:
I sit by the book table. Mike walks to the booktable and
begins picking up books. As he picks one up he says: See
this one? This an easy one. I can read this one. As he
continues, books slide off the table onto the floor. After
repeating this for about 15 books, he leaves the table.
He walks across two of the books he has dropped and makes
no attempt to pick any of them up.
Another check on observer effects was made by interviewing the
teacher who shared the room with Mrs. Saunders. Not only was this
teacher a close friend of Mrs. Saunders, but she was frequently in and
out of Mrs. Saunders' side of the room. The researcher asked the teacher
what kind of influence she thought the researcher's presence and
activities were having on Mrs. Saunders' behavior. The teacher's
I really don't think you have much influence on her. You're
really quiet when you're in there, and you've sort of made
it clear to the children because of your posture and your
behavior that you're not one of the helping teachers. The
only thing that I've noticed is that her hostess behavior
is up a little bit. I mean that I think she is poised
for company and for adult questioning more than she usually
is. But as far as her behavior with the kids, I can't tell
when you're in there and when you're not. Her teaching style
and her conversational tone with the kids don't change. I
have to go and look to see if you're there. I don't
feel like she's conscious of anything except your wish to
see the real situation.
Yet another procedure utilized to check on observer effects was to
compare children's and teacher's behavior in different situations.
Wilson (1977) suggested that the researcher compare the following:
a) what a subject says in response to a question; b) what
he says to other people; c) what he says in various situa-
tions; d) what he says at various times; e) what he actually
does; f) various nonverbal signals about the matter (for
example, body postures); and g) what those who are signifi-
cant to the person feel, say, and do about the matter.
By comparing speech messages, behavior, and use of materials, the researcher
could ascertain the degree to which observed events were unaffected or
natural. In addition to using her own fieldnotes to compare behaviors,
the researcher had access to fieldnotes recorded by an investigator who
had conducted a study in Mrs. Saunders' room during the previous school
year. Teacher-oriented observational and interview data from the second
study were strikingly similar to the data collected on the teacher in
the present study.
To summarize, the researcher took several steps to minimize and
to determine observer effects on studied behavior. A classroom was
selected in which observers were commonplace; the researcher spent 150
hours in the classroom; the researcher clearly established her role with
the teacher and children; the researcher compared data from various
sources; and the researcher interviewed another teacher to get her per-
ceptions of the researcher's impact on classroom life. The second
problem area concerned the effects of observer bias on data collection.
As Wilson (1977) pointed out, "No one, of course, enters a situation
a true tabula rasa previous experiences influence the scientist's
observation and thought" (p. 251). Observed phenomena are "selected
and filtered as well as interpreted and evaluated" (Schwartz & Schwartz,
1969, p. 90) by the investigator who enters the research site with
opinions, prejudices, and assumptions. Steps must be taken to deter-
mine the effect of observer subjectivity on recorded data.
Two features of the study which helped the researcher transcend
bias were the study's length and the nature of data collection. As
Bogdan and Biklen (1982) pointed out, "The researcher spends a con-
siderable amount of time in the empirical world laboriously collecting
and reviewing piles of data. The data that are collected provide
a much more detailed rendering of events than even the most creatively
prejudiced mind might have imagined prior to the study" (p. 42). In
addition to striving to record detailed, concrete, verbatim accounts of
observed behavior over a long period of time, the researcher included
subjective remarks and reflections in brackets. These subjective com-
ments served to remind the researcher of her prejudices and their
possible impact on the data collected. The process of confronting
personal biases is the main method of limiting their distorting effects.
Schwartz and Schwartz (1969) summarized the assumptions and conditions
of the active monitoring of researcher bias:
Implicit here is the assumption that bias is a universal
phenomenon; that the observer can and does know what his
biases are; and that, knowing what they are, he can, by
specifying them prevent distortions of his observations.
There are at least three conditions that need to be ful-
filled before this suggestion can be put into effect. The
observer must 1) be motivated to look for his biases;
2) look for them actively and, having come upon a bias,
explore its meaning and ramifications; and 3) look upon
the uncovering of his biases as a continuous process of
discovery--as an ongoing process to which there is no
end. (p. 103)
The researcher's biases grew out of her experiences as a public school
student, a teacher, and as an advanced graduate student specializing in
reading and working with preservice teachers. As these experiences are
related to the researcher's preparation for conducting the study, they
will be discussed in the section on researcher qualifications and biases.
The final problem inherent in participant observation, the re-
searcher's inability to observe and record all events related to
children's definitions of reading, was addressed in three ways. First,
by spending 150 hours observing in the classroom, the researcher was
confident that she had captured a thorough description of relevant
aspects of the phenomenon under investigation. Second, the teacher
served as a valuable informant, filling the researcher in on events
she had not directly observed. Third, the researcher's use of inter-
viewing and unobtrusive measures filled in gaps in observational data.
For instance, kindergarten teachers provided data which the researcher
could not have observed, and the children described some of their own
experiences and thoughts which otherwise would have been inaccessible
to the researcher. Regarding the use of participant observation and
interviewing, Becker and Geer (1969) noted, "There is considerable value
in using the strong points of one method to illuminate the shortcomings
of another" (p. 331). What are the shortcomings of interviewing?
McCall and Simmons (1969) summarized three primary threats to the
interpretability of interview data as follows: "1) the reactive effects
of the interview situation upon the received testimony; 2) distortions
in testimony; and 3) repertorial inabilities of the interviewee" (p. 104).
The main method of assessing the extent of these problems was by sup-
plementing interview data with observational data. According to Dean
and Whyte (1969), "the researcher is constantly relating the sentiments
expressed to the behavior he observes--or would expect to observe--in
the situation under discussion" (p. 109). The young children in par-
ticular had difficulties expressing themselves, especially in response
to abstract questions such as, "What is reading?" The researcher
evaluated a child's responses by considering them in light of what she
already knew about the child, by comparing them to responses of other
children, and by comparing them with the child's speech messages and
behaviors in other situations (Dean & Whyte, 1969). The two sources of
data provided a means whereby the researcher could evaluate the quality
of data collected by either method.
Making an Ethnographic Record
The data collected through observations, interviews, and unobtru-
sive measures were in the form of reading-related behavior, speech mes-
sages, and use of materials. The major portion of the data were recorded
in written form. Written records included fieldnotes, transcribed
interviews, and a research journal.
During observations the researcher took fieldnotes. The researcher's
goal was to record observed activity in as much detail as possible. She
attempted to record children's and teacher's language verbatim and their
actions in specific, concrete terms. These fieldnotes represented what
Spradley (1980) called a "condensed account" (p. 69) of what actually
occurred. That is, due to the speed and complexity of classroom activity,
the researcher typically recorded phrases and unconnected sentences
which would help her recall the details of observed events. Following
observation periods, such as when children went to a special class, the
researcher filled in details to create an "expanded account" (p. 70).
Still striving to use concrete descriptions and verbatim language, the
researcher indicated in the fieldnotes when she was directly quoting,
paraphrasing, or summarizing. Once the expanded account was completed,
the fieldnotes were typed into a formal protocol. Although the majority
of the fieldnotes represented classroom activity, a portion of each
fieldnote record contained the researcher's reactions to and impressions
of observed events. In recording these subjective comments the re-
searcher used brackets to separate them from the record of classroom
activity. Also in the fieldnote record the researcher sometimes in-
cluded questions she wanted to ask the teacher or the children. These
questions, like other subjective remarks, were bracketed.
Other data recorded in fieldnotes were descriptions of children's
completed assignments. The researcher often reproduced portions of ditto
sheets and workbook pages in the fieldnotes, particularly when children
were having difficulty with a task. Other diagrams recorded in field-
notes were classroom maps, chalkboard work, and examples of children's
writing, such as love notes and Christmas cards.
Formal interviews with the teacher, the children, and the teacher in
the other half of the room were tape recorded. All recorded interviews
were then transcribed and filed separately from fieldnote records.
Another written record was the research journal. "Like a diary,"
wrote Spradley (1980), "this journal will contain a record of experiences,
ideas, fears, mistakes, confusions, breakthroughs, and problems that
arise during fieldwork" (p. 71). In the journal the researcher kept a
record of the process of gaining entry to the classroom, dating and de-
scribing each experience with the teacher, the principal, and the re-
quired paperwork. Also in the journal were the researcher's musings
about her research goals and methods. Outlines for meetings with the
researcher's advisor were written in the journal as were the advisor's
comments. The researcher's reactions to observed events were recorded
in the fieldnotes rather than the journal in order to keep observations
and interpretations together. As the study progressed the researcher
recorded more subjective comments in the fieldnotes than in the journal.
It was found that recording these kinds of comments in the fieldnotes
was more convenient than switching to the journal.
Fieldnotes, transcribed interviews, and the research journal were
the written records of data collected in this study. As such they pro-
vided the substance for data analysis which was an ongoing process
comprised of several phases.
Analyzing Ethnographic Data
Just as the methodology was designed to be compatible with the
social-interaction perspective and suitable for the investigation of
children's definitions of reading, the choice of analytic strategies was
influenced by "the general purpose of the research, the nature of the
research problem or question, and the theoretical perspectives that in-
form the research problem and intrigue the researcher" (Goetz & LeCompte,
1981, p. 64). Goetz and LeCompte further noted that "the nature of the
problem or the way in which the research goal is defined is, of course,
the most significant of all design constraints" (p. 64). In this study
the researcher's goal was to discover definitions of reading constructed
by children in one classroom. To achieve this goal, many hours of
direct observation and interviewing took place. The researcher's task
in analyzing these data has been described as analogous to "putting a
jigsaw puzzle together" (Smith & Geoffrey, 1968, p. 15). The researcher
conducted a systematic search for order and understanding by "working
with data, organizing it, breaking it into manageable units, synthesizing
it, searching for patterns, discovering what is important and what is
to be learned, and deciding what you will tell others" (Bogdan & Biklen,
1982, p. 145). Spradley's (1980) model of data analysis was used to
guide the search for patterns in the hundreds of pages of data col-
lected in this study.
The process of data analysis was ongoing and consisted of four
phases. Each phase related to a type of question the researcher posed
and a type of observation she conducted. The types of questions and
observations, which are the basic unit of ethnographic inquiry, have
been discussed in a previous section. A cycle of questioning, collect-
ing, and analyzing was repeated throughout the study. The phases of
data analysis are described below:
1. Domain analysis was the first phase of the search for patterns.
During this phase the researcher identified categories of meaning, or
what Spradley (1980) called cultural domains. These categories were
discovered by reading the protocols with specific questions in mind.
Spradley identified nine types of questions which are useful in classi-
fying seemingly unique objects and events into categories. These ques-
tions include Are there kinds of things here? Are there places here?
Are there parts of things here? Are there results of things here? Are
there reasons for things? Are there uses for things? Are there ways to
do things? Are there stages in things? Are there characteristics of
things? These questions suggested categories but were not meant to
restrict the researcher's identification of categories. Examples of
some of the earliest domains were Kinds of Teacher Questions and
Places in Which Children Read. The search for domains continued
throughout the study. The researcher formulated structural questions
based on identified domains and made focused observations in order to
answer the questions. For example, related to the domain Kinds of
Teacher Questions the researcher asked, "What are all the kinds of
questions the teacher asks?" Observations focused on finding as many
answers to the question as the researcher could identify.
2. Taxonomic analysis was the second phase of data analysis.
During this phase the researcher analyzed domains to find out how they
were organized. A taxonomy reveals the organization of domains by
showing the relationships among the terms inside the domain. Taxonomic
analysis also helps the researcher to relate domains to one another.
For example, the domain Kinds of Questions was a very large domain
which included teacher questions and student questions. Within each of
the two levels of student and teacher questions were still more levels.
Within the domain of student questions, for instance, were student-
student questions and student-teacher questions. Within these levels
were more levels, such as technical assistance questions ("How do you
spell 'who'?"), directions questions ("What do you do on this page?"),
and socializing questions ("Wanna play school?"). In developing
taxonomies the researcher attempted to fit together the pieces of the
scene (domains) already identified. Associated with taxonomic analysis
were contrast questions which led to selected observations.
3. Componential analysis is a search for the characteristics of
identified domains and of the terms within domains. If an object or
event has meaning in the setting, it has certain attributes regularly
associated with it. For instance, if directions questions were a type
of student-student question in the classroom, there were certain charac-
teristics which defined this kind of question. The goal of componential
analysis was to determine whether identified domains and terms within
domains were distinct elements in the setting under investigation.
4. Theme analysis involved the search for a theme which tied
together the identified parts of the scene. During this phase the
researcher looked for meanings which recurred across domains. Although
a theme may not have applied to all parts of the scene, it had to have
a high degree of generality and serve to link at least several
The analytic strategies utilized in this study constituted a
systematic, rigorous organizational process. Data were analyzed to
identify categories of objects and events related to children's defini-
tions of reading. Further analysis revealed relationships among iden-
tified categories and the relationship of the parts to the whole class-
room scene. Throughout all phases of data analysis, questions emerged
which served to direct the researcher's observations of the scene. The
interactive nature of data collection and analysis is a fundamental
principle of the research model utilized in this study. In the final
two sections of this chapter these two topics related to data collection
and analysis will be discussed: (a) researcher qualifications and
biases and (b) validity of the researcher's findings.
Researcher Qualifications and Biases
Since the qualitative researcher is the key research instrument,
qualifications and biases which may influence data collection are impor-
tant to consider. Wolcott (1976) identified several criteria for
judging the adequacy of an ethnographer. Because the research approach
utilized in this study "borrows generously from ethnographic techniques"
(Wolcott, 1976, p. 30) and was utilized to identify participant per-
spectives, as would an ethnographer, Wolcott's list of qualifications
were considered to be relevant. According to Wolcott, the researcher
must be a "sensitive and perceptive observer, at once sympathetic,
skeptical, objective, and inordinately curious." Furthermore, the
researcher requires "physical stamina, emotional stability, and per-
sonal flexibility" as well as "the skills of the story-teller and writer"
(p. 28). The most important qualification is experience conducting
fieldwork. Although cross-cultural fieldwork is considered by many to
be a mandatory experience for an ethnographer, Wolcott suggested that
one could substitute experience doing microethnography in "education-
ally-relevant events and settings" (p. 29) for cross-cultural research.
Although the researcher's flexibility and writing skill will have to be
judged by the reader, the following experiences related to Wolcott's
criteria are listed below:
1. The researcher was a classroom teacher for four years, two years
at the secondary level and two years at the elementary level. At the
secondary level the researcher taught reading and language arts.
2. The researcher earned an M.Ed. in the area of reading.
3. The researcher has completed coursework for a Ph.D. in cur-
riculum and instruction, including courses in elementary curriculum and
curriculum development. Specializing in reading, the researcher com-
pleted extensive coursework in this area.
4. The researcher has taken two courses which provided a theoreti-
cal and practical background in qualitative research. In addition to
the readings required by these courses, the researcher has read exten-
sively in the area of qualitative research foundations and methods as
indicated by entries in the reference list of this report.
5. The researcher has completed two qualitative studies in elemen-
tary school classrooms. A report of each study was written, and one
was presented at a national conference.
6. The researcher has gained additional experience as an elemen-
tary classroom observer by supervising student teachers over a two-year
7. The researcher has worked to develop her writing skills by
preparing manuscripts for publication and for presentation at state,
regional, and national conferences.
In addition to meeting certain criteria for conducting qualitative
research Wolcott suggested that the fieldworker "needs to grapple with
his own 'underlying assumptions' and recognize the kinds of evi-
dence he is most attracted to in building his account" (p. 27). Schwartz
and Jacobs (1979) referred to this process as "true confessions" (p. 58).
In recognition of the potential impact of the researcher's values and
biases on the nature of the data collected and on project outcomes, the
following list of relevant beliefs is provided. By listing these be-
liefs the researcher demonstrates awareness of them and provides the
reader with a basis for evaluation of the study (Ross, 1978).
1. The researcher objects to the widespread practice of teaching
reading as a sequence of subskills isolated from the purposeful reading
of various types of text. Related to this is the researcher's concern
over the domination of reading instruction by the teacher's manuals of
the basal series. Although the researcher believes that there is nothing
inherently wrong with basal readers, she would prefer to see instruction
reflect closer attention to children's language and experiences.
2. The researcher is particularly concerned about the kinds of
reading instruction provided to low-achieving children. Typically these
children receive repeated drill on isolated reading subskills. It is
believed that often this instruction does not serve them well.
3. The researcher believes that classrooms are complex, dynamic,
multi-dimensional environments in which children and teachers influence
and shape one another's behaviors.
4. The researcher assumes that children's perceptions of reading
are influenced by numerous factors, and their perceptions may not always
be congruent with the teacher's perceptions.
In qualitative research validity is a central concern (Erickson,
1979; Hymes, 1982; Rist, 1977). According to LeCompte and Goetz (1982),
"Establishing validity requires determining the extent to which conclu-
sions effectively represent empirical reality and assessing whether
constructs devised by researchers represent or measure the categories
of human experience that occur" (p. 32). That is, the researcher strives
to "improve the fit between the model and the reality" (Lancy, 1978,
p. 125). Some of the steps taken to ensure the validity of the study's
findings have already been discussed. For instance, the long period of
data collection enabled the researcher to become familiar with the
scene and its participants. Also, the use of several methods of data
collection provided opportunities to compare data and to probe deeply
into participants' perspectives. Another procedure the researcher
utilized to contribute to the validity of the findings was to search
for negative examples of hypothesized components of the model of
children's perceptions of reading. Becker (1970) summarized this
After constructing a model specifying the relationships
among various elements of this part of the organization,
the observer seeks greater accuracy by successively refining
the model to take account of evidence which does not fit
his previous formulation; by searching for negative cases
(items of evidence which run counter to the relationships
hypothesized in the model) which might force such revision;
and by searching intensively for the interconnections in
vivo of the various elements he has conceptualized from-his
data. (p. 34)
Yet another procedure for establishing the validity of the researcher's
findings was to discuss them with some of the classroom participants.
By sharing findings with the teacher, the researcher received valuable
feedback on her interpretations of participants' perspectives.
In the next two chapters the researcher's findings are described
and discussed. Descriptions of the children's definitions of reading
and evidence to support the existence of these definitions are presented
in the first of the two findings chapters. In the second chapter
children's definitions are explained in terms of a factor which emerged
as a powerful force in definition construction, the teacher's classroom
CHILDREN'S DEFINITIONS OF READING
The goal of this study was to uncover the definitions of reading
constructed by first-grade children in the low and high reading groups
in one classroom. As previously discussed, the researcher adopted a
social-interaction perspective wherein it was assumed that individuals
construct definitions of reading through their interactions in social
contexts. In this study the researcher focused on interactions which
took place within one classroom throughout the school day. More speci-
fically, observations were focused on children's speech messages about
reading, their reading-related behavior, and their use of reading
materials. These kinds of concrete phenomena served as indicators of
children's definitions of reading.
The collected data were analyzed into domains or categories accord-
ing to similarities among recorded events. Domains which proved to be
particularly helpful in revealing reading definitions included Kinds of
Reading Miscues, Kinds of Things Children Do with Chosen Books, Kinds
of Statements Children Make About Books and Assignments, Kinds of State-
ments the Teacher Makes About Children's Reading, and Kinds of State-
ments Kindergarten Teachers Make About Children's Reading. Taxonomies
were constructed by integrating data from different domains. That is,
data which indicated a particular definition of reading were drawn from
across domains and organized into new domains which represented defini-
tions of reading. Taxonomies were also constructed to represent the
reading-related behavior of individual children. The purpose of this
analytic exercise was to verify the existence of identified definitions
through careful examination of the behavior of each child who seemed to
use a particular definition. As children's definitions are described,
data from the taxonomies will be provided to support and illustrate the
definitions. It should be noted that illustrations included were
selected from among many examples and do not represent the sole indi-
cators of a particular definition.
Among the studied children in this classroom, those in the highest
and lowest reading groups, there were six definitions of reading (see
Table 1). No single definition was shared by all children. Some
definitions were peculiar to one group while other definitions were
shared by members of both groups. Most children constructed and utilized
more than one definition of reading.
The first three definitions of reading--reading is saying words
correctly, reading is schoolwork, reading is a source of status--were
common among low group children. The schoolwork and status definitions
were also utilized by two girls in the high group. What did these three
definitions have in common? Readers utilizing these definitions seemed
to view reading as an externally imposed task. While this does not mean
that the readers did not experience some pleasure associated with
reading, the pleasure was related to pronouncing the words correctly
(reading is saying words), getting the job done (reading is schoolwork),
or gaining recognition (reading is a source of status). Children
O a Nr- OU. L. E
U) )i N -- U <) E
(0 *r- 0 i- *r- C 3 0
'--0-I .-J- C 0 I---
C M.0 U
U CU r-r-
r- 0 ) 0 0 S.
LU -) X: = I-
*r- 0 *r-
"0 in 4J
*- o *r-
n S- m
0 2 U
r 3: O v
r 0 0)
U) > o
utilizing these three definitions were not likely to engage in private
experiences with books or to choose to spend time with books under
free-choice conditions. These definitions were associated with the
view of reading as a required task.
The next three definitions--reading is a way to learn things,
reading is a private pleasure, reading is a social activity--were
utilized almost exclusively by the high group girls. Unlike the first
three definitions, these reflected a view of reading as a personally
meaningful activity. Children utilizing these definitions sought out
books during free-choice periods and were, in general, busier with
books than children utilizing the first three definitions.
Reading Is Saying Words Correctly
Perhaps the clearest definition of reading to identify in this
classroom was that reading is seeing and saying words. This definition
was constructed and utilized by all children in the lowest reading
group. Not only did the definition serve to guide their reading be-
havior during reading group time, but it influenced the nature of their
reading in other classroom contexts. Although high group children
referred to the importance of learning words, there was a clear dis-
tinction between the two groups' words-based definition of reading.
While high group children viewed knowing words as part of a larger
process, the low group children viewed saying the words correctly as
an end in itself. For these children reading was an oral performance
involving calling out words. Often the meaning of a word or the sense
of a series of words was ignored. Saying the words was more important
to low group children than finding meaning in written language.
During reading group sessions, the majority of low group children's
talk was in the form of answers to the teacher's questions. However,
children's spontaneous questions and comments frequently focused on
saying words. A standard part of the reading lesson involved reading
cards on which the teacher had printed words from the pre-primer. Much
of the children's talk about reading was related to words on cards.
For instance, children often asked if they could say the words. Such
statements were of the form, "We gonna do our words?" or "I wish we
could say the words. I wish we could say the words every day." The
children also asked the teacher if they could get new words and asked
one another if they had a particular word, as in the following example:
Lizzie says to each child in the group: Don't you got
"no"? (She holds up a card with "no" printed on it.)
Sharon: Everybody got "no"!
Lizzie: Let me see something 'cause I might have two
Sharon: Teacher, I don't have "not." (She repeats
Teacher to Lizzie: I think you're the only one who
Teacher asks group: How do you spell "not"?
Lizzie's attention to words even extended to the researcher's activi-
ties. During reading group she observed the researcher writing and
remarked to the other children, "She's doing our words. She's writing
our words down."
Low group children also talked about the words they had at home.
Their comments included, "I got these at home," "My mamma gots lots of
words," "My mamma showed me words on cards like this, and I read them."
The children also talked about how many cards they had, with such com-
ments as, "I got a lot of words! My pin (paper clip) can't fit on!"
These children exhibited other behaviors during their reading group
time which indicated their concern for saying words and for saying them
correctly. For example, they asked questions about words, such as,
"Ain't this 'hide'?" and "What is this word?" In addition, some of the
children were quick to criticize those who said words incorrectly, as
the following fieldnote excerpt illustrates:
The group reads a sentence in unison: This is the park.
(Some voices say "the" instead of "this.")
Lizzie: Teacher, they don't do it right!
Sharon: Uh-huh! This is the park!
Lizzie: Not the first time they didn't!
Teacher: We're all learning to read here Lizzie, and
people do it at different times.
Yet another example of the children's words-based definition occurred
when the teacher announced that the group would skip a page in their
Teacher: I'm going to skip 20. You already know
"I am." We can skip this. Turn to 21 and fold it
Sharon: Why not this page? (She looks at a page with
an airplane on it and no print.) Oh, it ain't got no
words on it.
The children also indicated their "saying words" definition of reading
by calling out words before the teacher arrived at the reading group
table. One child would pick up the pile of cards and show a card to
each child at the table. The group cooperated as the student held up
a card and asked, "What's this say?" When a child hesitated on a word,
others eagerly called it out.
The low group children's behavior outside of the reading group pro-
vided further evidence that they defined reading as saying words
correctly. These children neither chose books as often nor spent as
much time with books as did the children in the high group. This was
not surprising given their definition of reading. Since they defined
reading as saying the words correctly, and since they could not say many
of the words they encountered, they did not view reading books as an
attractive free-time activity. While the children chose books less often
than the high group children, their interactions with print and their
comments about reading indicated a words-based definition.
Several of the children read individual words aloud at their desks
much as they read word cards during reading group. In the following
fieldnote excerpt, one child read her cards and then showed them to the
child next to her:
During sharing time, Lizzie returns to the room from her
remedial tutoring session. She sits down and looks at each
of her three word cards. She whispers the three words:
"here," "yes," "not." Then she turns to Melissa and shows
the cards one at a time. Melissa stares blankly.
In a similar incident, Lizzie and Sharon read a list of rhyming words
on a commercially produced chart:
Lizzie: Sharon, we're gonna say some of these, okay?
(She points to each word and pronounces it as Sharon
Lizzie's interactions with books were guided by her words-based
definition of reading. For instance, she was observed reciting words
from the pre-primer while turning the pages of trade books from the book
table. She clearly enunciated words she knew, despite the fact that
these words were not to be found in the trade books. In the following
fieldnote excerpts Lizzie demonstrates her definition of reading as an oral
Lizzie holds a book. She opens it and says to the re-
searcher: This is a working book. A Big Bird working
book. See, you read and then circle. She points to words
in the book while saying: Bill-is-not-here. (These words
are not in this book.)
In a similar incident, Lizzie called out a number of words from the pre-
On Mike's desk there are word cards which belong to a
rhyming game. Lizzie walks over, touches "run" and says:
Run. Run, run, said Lad. Run, run, said Jill. Run, run,
said Ben. Run, run, said Bill.
Other children in the low group utilized a words-based definition
of reading. Mike, for instance, would only read out loud, standing by
an adult. The teacher commented, "I think Mike thinks he isn't reading
unless he's saying the words aloud to someone." As he read orally, Mike
concentrated on saying words rather than making sense of the words he
was saying. On one occasion, he stood next to the researcher and read
from the book, Let's Go, Dear Dragon. He misread a number of words and
never corrected himself:
Mike: Get want. Get want. (Get out. Get out.) No
one can gets (guess) where you are. That it not go.
(That is not good.)
Jason's words-based definition of reading and his confusion about
written language were reflected in his experience writing and reading
a Christmas card. In this incident the researcher may have (uninten-
tionally) indicated to Jason that there was a problem with his card:
Jason to researcher: I'll read you my story. Merry
Christmas Mom Dad I love are I can buy my. (Jason looks
up at the researcher.)
Researcher: What do you want to say?
Jason: I don't know.
Researcher: You were probably thinking of something
you wanted to say to your Mom and Dad.
Jason: I don't know. I'll erase from here. (He erases
"are," "but," and "my.")
Researcher: What do you want to say to your Mom and Dad?
Jason: I could say Happy Christmas. How do you spell
Happy? (He writes the word "Happy" in front of the word
"Merry," closes the card, and draws a Christmas tree on
After having copied "Merry Christmas Mom (and) Dad" from the blackboard,
Jason wrote and then read a series of words which did not make sense.
When the researcher asked, "What do you want to say?" Jason seemed to
realize that something was wrong, but he did not know how to correct his
error. Another incident involving Jason further demonstrated his defini-
tion of reading:
Jason to researcher: I only have two smilies left. This
and this. (He taps his reading workbook and his math book.)
As it is Friday, the day students read a book for a smilie,
the researcher asks: What book will you read today?
Jason: Uh, prob'ly this one. (He touches a coloring book
which supplements the pre-primer.) Maybe this. (He reaches
for a trade book which also goes with the pre-primer.) This
is the new one.
Researcher: When you get ready to read it, will you get
me so I can hear you?
Jason: Well, I can't even read it.
Researcher: You can't?
Jason: No, you have to teach me. Read it to me, then I
can read it back to you.
Jason seemed to be equating reading with saying the words correctly--
something he could only do by repeating the words after an adult had
During whole-class phonics activities the children often demonstrated
their words-based definition of reading. When working in the phonics
books, the teacher focused attention on sounds of letters and correct
word-calling. Sharon's behavior reflected this child's definition of
The teacher calls on Sharon, who has been wildly waving her
arm to be allowed to read a sentence. Robin, who sits next
to Sharon, prompts her on each word. Robin whispers the
word, Sharon calls it out, loudly, and looks triumphantly at
the teacher after each word. After this sentence, Sharon
points to the next sentence and quietly says: Jan will fish
the pan said the ___ The sentence says: Jan will fix the
rip in the She waits for Robin to fill in the blank,
and then copies the word.
Sharon's concern for saying the words correctly was shared by Jason.
Jason, who frequently complained, "I can't read this thing!" became
excited when he recognized his name in the phonics book. He called out,
"Jason! Mrs. Saunders! That's Jason! My name!" The teacher responded,
"Yes, it is. Jason's going to read it because it begins with his name."
Immediately Jason claimed, "I can't!" and then again pointed out his
name, the one word he could read in the sentence: "Mrs. Saunders, see?
The children's responses to interview questions designed to uncover
their definitions of reading indicated that they defined reading as
saying words correctly. For example, when asked, "What kinds of read-
ing do you do?" several children responded by reciting words from the
Researcher: Tell me something that you read.
Sharon: All right. I can read, "Jill and Nan hide at the
park ." and I can read a book that says, "Ben and Ted
will run to the park."
When asked other questions about her reading, Sharon responded by re-
citing story lines in a slow, halting voice as a beginning decoder
might. After she announced that her mother had taught her to read,
the following exchange took place:
Researcher: How did your mamma teach you to read? What
did she do?
Sharon: She told me to read, to say, "Boy meet girl," and
I say, "Boy meet girl," and then she told me to say the
rest of it. .
Sharon's remarks provided further evidence of her view of reading as
saying the words.
The children's descriptions of the best readers in the class also
reflected their prevailing definition of reading. Children said they
knew who the best readers were because they saw and/or heard them read.
Good readers were people who "do the sounds" or "sound out the hard
words." When asked, "What do I have to be doing when I'm reading?"
Richard responded, "You gotta be talking."
Finally, children made numerous remarks during the day which pro-
vided evidence for the existence of a words-based definition of reading.
For instance, their comments about assignments were sometimes revealing:
Researcher asks Richard: Tell me, Richard, why does
Mrs. Saunders make you do this page?
Lizzie overhears and answers: Cuz we learn.
Researcher: What do you learn?
Lizzie: To read--see (she points), this says "big" and
this says "little."
Richard points to "yes": And this.
Several days later Richard looked at the researcher's fieldnotes,
pointed to the words "go," "I," "Jason," "Lad," and "is" and read each
as he pointed. He then said, "You have Jason a lot. And there's Mike.
I know a lot of words you know." The researcher asked, "What do you
know?" Richard replied, "Duck, cat, get, not, Ted, Ben, Bill, yes, run,
duck--did I already say that?" Richard again demonstrated his words-
based definition when he tried to read the first words in a trade book--
the small print which presented copyright information. After staring
at the print for several seconds, he took the book to Mike and pointed
to the words, saying, "What does this say?" Mike replied, "You don't
gotta read that part."
Children's remarks about other children's behavior also revealed
their definitions of reading, as illustrated in the following incident:
Jason is kneeling at the book table with The Golden Goose
open in front of him. He reads: You-do-you-do-not--
Bonnie interrupts: You don't know how to read! (She
looks at the page and reads aloud.) You do not like--
(Jason closes the book and holds it close to his face.)
Lizzie to Jason: You can't read those words.
The low group children's reading-related behavior was guided by
a word-based definition of reading. In this classroom the children
demonstrated an acute awareness of words and a great concern for saying
them correctly. Their definition was reflected in their behavior
throughout the school day, and may have restricted these children in
the other definitions they could construct. That is, if a child de-
fined reading as saying words correctly, and if he/she was not able to
read many words, the child was not likely to construct a definition of
reading as, for instance, a private pleasure, as did most of the high
group children. In this classroom, low group children did not con-
struct such definitions.
The high group children were clearly aware of words, but their
reading-related behavior was guided by other definitions of reading.
In response to interview questions, the high group children pointed out
the importance of sounding out words and practicing words in order to
be a good reader. For example, when asked, "What do people have to do
to be good readers?" Robin replied, "Well, they can practice and study
their words a lot." To the same question Jane replied, "They have to
sound out the words and get their parents to help them read the words."
However, high group children also talked about a variety of purposes and
reasons for reading. Knowing words seemed to be part of the machinery
of reading for fun, reading for one's job, reading to learn, and reading
to identify things in the environment. While the words-based definition
helped explain much of the low group's reading behavior, it did not have
the same explanatory power for the behavior of the high group children.
Reading Is Schoolwork
Most of the low group children and two of the high group children
defined reading as schoolwork. This definition was associated with
reading-related comments and behaviors which seemed to announce, "We
do it because we're supposed to, but we'd rather be having fun." For
these children, reading was work, just as math and handwriting were work.
Reading was just another "smilie" to be completed in order to move on
to free-choice activities.
An early indication of the work definition was that many children
did not choose books during free-choice periods. Although these children
sometimes chose books when given the choice between a book or drawing
paper, they did not select books when given a wider choice of activities.
Further, when some children discovered that instead of choosing a book
during attendance and lunch count, they could begin their morning
smilies, they abandoned books altogether. Jason and Tommy exemplified
this behavior. The two boys began their morning work as soon as they
could in order to finish it as early as possible. Tommy stood at his
desk as he did his smilies. It seemed that he thought standing up gave
him quicker access to assistance and hence completion of his assignments.
Frequently he asked whoever was closest to him at the moment, "What's
this word?" and "How do you do this?" Often Tommy exclaimed, "Finished!"
as he closed his folder, slapped his pencil down on the desk, and
quickly moved on to play games on the floor.
Jason's behavior was similar to Tommy's, but Jason was more vocal
about his feelings toward work. He excitedly spread the news to his
classmates when he found that there would be no smilies one morning.
On another occasion, when the researcher asked why the teacher had given
him a reading worksheet, he replied, "To do our work." On another
occasion he spoke of several subject areas as if they all amounted to
the same thing--work. His comments were prompted by the researcher
asking him what kinds of reading he did in school. He answered by
listing the smilies he typically did, including math. To Jason school
subjects seemed to be interchangeable. As with Tommy, Jason preferred
to be on the floor with games and puzzles rather than at his desk
"working." According to the two boys' kindergarten teachers, they
behaved similarly during the previous school year. Neither would
choose books during free time; both enjoyed playing with toys and
games. For these boys, school seemed to be defined in terms of work
and play. Reading was clearly perceived within the work component of
Other evidence to support the existence of a "reading is work"
definition came from interviews with the children. All of the low group
children talked about reading as something people have to learn how to
do. When asked, "Why do people read?" they gave the following kinds
'Cause they want to learn how to read.
So when you grow up you can know how to read.
'Cause they don't know what things say and they can learn
how to read.
The low group children reported that people read to learn how to read.
In other words, reading is work one does when one is young in order to
be able to do something (read) when one is older. Some children used
the word "work" when talking about reading. For instance, when the re-
searcher asked Joseph, "Why isn't anybody reading now?" he answered,
"'Cause they finished their work." Tommy said several times that the
way to become a better reader was to work. Having discovered that many
children defined reading as work shed light on the behavior of low group
children. The definition also helped to explain why the children did
not favor books during free-choice periods.
Low group children were not alone in defining reading as school-
work. Two of the high group children also utilized this definition.
Tracey and Sally stood out among high group children in their apparent
lack of interest in books. Once it became clear that the schoolwork
definition explained some of the other children's behavior, the re-
searcher considered the possibility that it might apply to Tracey and
Sally. Indeed, this definition was clearly one of several constructed
by the two girls. As with the low group children, these girls did not
often choose books during free-choice times. During periods when chil-
dren were expected to read, the girls would busy themselves with books.
For instance, they would go to the book table, select several books,
and return to their seats. On one occasion Tracey put the books inside
her desk and turned to the boy next to her to say, "Go get your folder."
The next morning she returned the books to the book table. Sally en-
gaged in a similar activity of carrying four books to her desk and then
quietly reciting, "Eeny, meeny, miny, mo" over them until the teacher
began the morning's lesson.
Another indication of the girls' definition of reading was the kind
of books they selected from the classroom book table and bookshelves.
Both girls often chose basal readers and teachers' editions of basals.
These books, some of them more than fifteen years old, had been placed
among trade books, encyclopedias, dictionaries, and magazines on the
table and in the shelves. Tracey and Sally chose them but were not ob-
served reading them. Sally found an old basal reader in the library and
checked it out. The girls' book choices suggested that reading was
associated with teacher-directed lessons and teacher-selected materials.
Reading was work one did in school with teachers.
Other evidence of the girls' definition of reading was their be-
havior during the required 30-minute reading period after lunch. During
this time the children were allowed to interact with one another but
only if books were the focus of their interactions. Whereas some chil-
dren chose to read at their desks, Sally and Tracey were most often
involved with other children during this time. A commonly observed
activity was playing school:
Sally pulls her chair up to Donna and Shelley.
Sally: We can play school.
Donna: I'm the teacher. I'm reading the book.
Sally opens Sun and Shadow (a basal reader) and reads
Shelley: I'm going to read you a story.
Teacher: You may choose now what you would like to play.
Sally and Shelley drop their books on Donna's lap and
dash to the playhouse.
Donna looks up: Gosh, I got all the books.
Not only did Sally define the activity as schoolwork, but she read from
a school book. When Sally was asked how people learn to read, she
talked about starting in "the first book" and going through "the dif-
ferent levels." Again, she indicated a view of reading as work one did
in school. When the work period was over, as in the school episode de-
scribed above, Sally dumped her book and headed off to play. For Sally
and Tracey and for the low group children, a schoolwork definition of
reading influenced their reading-related activities. The same set of
children shared the next definition of reading.
Reading Is a Source of Status
Analysis of the kinds of remarks children made about books and
reading in this classroom suggested a third actively utilized definition
of reading. Many of children's comments could be characterized as "look
at me" statements, or statements which served to draw the attention of
others to the speaker's activities. The following are examples of this
frequently observed kind of remark:
Lizzie: Mrs. Saunders, wanna see my book?
Joseph: This is a good story. I got so many books!
Richard: Anybody can read these old books. See? Run,
Richard: I read that book right here and this one.
Mike: Man, I can't find no book! Oh, yeh! Just
the book I been looking' for!
Lizzie: I got me a great book!
Richard to Jason: She (teacher) said I did good!
Children always clearly announced these statements to someone or to a
group. The public nature of these self-asserting comments suggested
that not only did some children recognize the great value placed on
reading, but that they defined reading as a source of status. Inter-
actions among some of the low group boys further supported the existence
and utilization of the status definition:
Mike: Look, two new books!
Joseph: Oooo, could I have this one?
Richard: Here the good books, man!
These boys often gathered noisily at the book table. They exclaimed,
demanded, and announced their intentions for all to hear. Often they
would leave the table empty-handed, suggesting that the performance was
more important than the books themselves. The following remarks were
Mike: I'm gonna get me a good better book today!
Richard to Bonnie: Uh-uh! You ain't taking these! (He
grabs several books from her.)
Richard looks at pictures in a book: My daddy has more
cars than this. He has 'em all around the house. They
go all over the brick wall.
Mike: This was a fun book! I know what all these books
Mike's reading behavior clearly reflected the influence of the
status definition. Not only did he always read aloud, but always in
the presence of someone, usually an adult. When he read, he frequently
looked at the person or people near him as if to be sure they were
attending to him. Mike was one of the few children who regularly ap-
proached the researcher. Often he wanted to read aloud. At other times
he confidently explained things to the researcher, as in the following
Mike: Look at this book about animals! Look at these
dinosaurs here! (He holds up the books.) Finally! I
found a good book. This is football. (It's a magazine.)
Just sign it, and you go to football practice. Right
here, see? (He points to a coupon.) Name, order, sign,
care. (The words are name, organization, state, city.)
You like swimming? Look--just sign up and you go swimming!
You like soccer? Just sign and you go.
Researcher sits by the book table. Mike walks to the table
and begins picking up books. As he picks one up he says:
See this one? This is an easy one. I can read this one.
Mike's reading-related behavior and comments indicated that he was
attuned to the recognition and status given to readers in this classroom.
He went so far as to claim that he was the best reader in the class,
the only low group child to do so. Reading for Mike was a public act.
It was associated with asserting his superior position among peers--a
means of achieving high status in the classroom community.
Although Joseph and Richard also utilized a status definition of
reading, they exhibited the definition differently from Mike. Joseph
and Richard were unskilled word-callers. They knew many fewer words
than Mike and consequently did not often loudly broadcast their oral
reading, as did Mike. Nevertheless, the boys demonstrated the status
definition in other behaviors. Both frequently grabbed books from other
children, for no apparent reason other than to assert their superior
positions as the ones who had the desirable books. They also were
observed to carry piles of books to their desks, where they would often
sit unopened while the boys played with friends at neighboring desks.
Evidently, the public display of choosing books from the book table
was more important than anything they might do with the books once they
arrived at their seats. This is not to say that the boys were never
actively engaged with books at their seats. But even in these situa-
tions reading often became a competitive enterprise, as this excerpt
from fieldnotes illustrates:
Richard turns the pages of Clifford's Riddles.
Richard to the researcher: I bet this dog will pull the
Researcher: He will?
Richard: He's really big, see? (He shows a picture at the
beginning of the story.)
Joseph looks over and says: That ain't the same dog!
Richard: Uh-huh! Watch, he's gonna change. Want me to
show you the whole book?
Low group children were not alone in defining reading as a source
of status. Tracey and Sally clearly used this definition to guide their
behavior in public contexts. As discussed earlier, these two girls did
not often choose to read during free-choice periods. However, when the
teacher announced that people who finished reading 10 books and re-
cording the titles would get a sticker and a candy cane, Tracey and
Sally became very busy with books. Sally immediately took three books
from the book table and, returning to her seat, rapidly turned through
the pages of each book. Although she turned each page, she could not
possibly have read the pages as quickly as she turned them. Soon Sally
was standing, booklist in hand, by the teacher. After showing her list
to the teacher, she walked to the researcher and said, "Ms. Bondy, I
was the first one to finish the whole page." The researcher responded,
"All 10 books?" Sally replied, "Yeh, the first one."
During this time, Tracey sat at her desk, a pile of books in one
corner. She hastily turned pages and recorded titles. Tracey appeared
to have orchestrated a more efficient system than Sally, as she had
other children transporting books for her:
Robin to Tracey: You should read this poem book. It's
got lots of easy poems.
Tracey: I'll read it if you put this one back.
Lynn presents a book to Tracey: Tracey, read this book,
it's cinch. It's about Stanley, and you'll love it.
Tracey looks up: No, I don't want to.
As Tracey added books to her list, Sally worked on her second list.
Lynn provided periodic news bulletins:
Lynn to Sally: Tracey only has three more books and she
gots them all.
Lynn to Mrs. Saunders: Mrs. Saunders, Tracey only has
to write the book, and she'll be done.
Having recorded the tenth title, Tracey closed her folder and dashed to
Mrs. Saunders to announce, "I'm finished." She chose a red candy cane
from the bag and promptly displayed it to Sally and Robin.
By bringing books to sharing time, these girls and other children
revealed their use of reading to gain status. In front of the whole
class, children displayed books they had brought from home. Not all
children were operating under the status definition; some children were
clearly more attuned to reading as a pleasurable experience or as a
way to learn things, as will be discussed later. However, Tracey and
Sally were guided by their definition of reading as a high-status
activity. In public settings which provided opportunities for recogni-
tion, these girls, like the low group children, engaged in reading and
reading-related activities. Tracey's response to the question, "What
is reading?" further clarified the status definition. She said, "Like
you look at a book, then you write it down in the folder, and then the
next day you could read another book." Only by recording book titles
could a student complete the list in the folder and become the object of
teacher and peer attention. Reading was viewed as an activity done with
the promise of praise, reward, and public recognition.
Reading Is a Way to Learn Things
When interviewed, high group children talked about many more pur-
poses and reasons for reading than low group children. These girls
pointed out that reading was useful to people for a range of reasons.
One reason for reading was to find out about things in one's environ-
ment. In the following excerpts from interviews, Meg, Robin, and Ellen
referred to reading as a way to learn things:
Researcher: Tell me what kinds of things people read in
Meg: Um, school books.
Researcher: What do you mean when you say school books?
Meg: I mean like reading group books.
Researcher: Oh, so sometimes they read those. What else?
Meg: Um, their own books they bring from home.
Researcher: Other things?
Meg: Yes. Sometimes they read like that box right there:
"The Scott Land Clean of Fine Protection Papers."
Researcher: So sometimes people read things that are
written on boxes. Other things?
Meg: Yes, like names, over there. They might read all the
names on the wall there and then make a pretend book with
all the names. Like this book right here. (She picks it up
and reads it.)
Researcher: Why do you think people read, Meg?
Meg: Well, sometimes they read for working. And sometimes
they read for playing. And sometimes they read really for
teaching people. Sometimes they read to find out things,
like the pages in math and the pages in handwriting. And
sometimes they read to find the days, like here. (She
touches the calendar.) Like Monday, Tuesday, Wednesday,
Thursday, Friday, Saturday, Sunday.
Researcher: What kinds of things do you read, Ellen?
Ellen: Well, in reading group I have to read level six
book. Sometimes I read books I want to when it's free
Researcher: Are there any other things you read?
Ellen: At home I read a lot. My mom and me read together
when I go to bed.
Researcher: Other things?
Ellen: Well, when I'm doing my workbook, I have to read to
find out the answer. And right now. I could read off the