I

«H ,5 I III III III III III I III II IIII II III

LIBRARY

Michigan State

University

 

This is to certify that the

thesis entitled

THE CLINICAL PROBLEM SOLVING BEHAVIOR OF CLASSROOM
TEACHERS AS THEY DIAGNOSE CHILDRENS' READING
PERFORMANCE IN EXPERIMENTAL AND CLASSROOM SITUATIONS

presented by

DORON GIL

has been accepted towards fulfillment
of the requirements for

Ph.D. Eduaation

degree in

 

 

    

Major professor

11.1.79
Date

 

0-7639

 

THE CLINICAL PROBLEM SOLVING BEHAVIOR OF CLASSROOM TEACHERS
AS THEY DIAGNOSE CHILDREN'S READING PERFORMANCE

IN EXPERIMENTAL AND CLASSROOM SITUATIONS

BY

Doron Gil

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree

DOCTOR OF PHILOSOPHY
Department of Student Teaching and Professional Development

1979

ABSTRACT

THE CLINICAL PROBLEM SOLVING BEHAVIOR OF CLASSROOM TEACHERS
AS THEY DIAGNOSE CHILDREN'S READING PERFORMANCE
IN EXPERIMENTAL AND CLASSROOM SITUATIONS

BY

Doron Gil

The purpose of this study was to investigate the
clinical problem solving behavior of classroom teachers in
reading diagnosis.

The literature on reading instruction views classroom
diagnosis as an essential part of the total reading instruction.
Most researchers believe classroom teachers have the responsi-
bility to conduct necessary diagnoses and provide the
appropriate remediations for students with reading defi-
ciencies. No empirical evidence exists, however, to show how
capable classroom teachers are of performing these functions.
Also lacking from the literature is a description of the
cognitive processes teachers employ as they attempt to
diagnose students' reading difficulties.

There were two objectives to this study: (1) to
investigate the clinical problem solving behavior of class-
room teachers, in both experimental and classroom situations,
as they gather data about children's reading difficulties to

arrive at diagnostic judgments about the children's

Doron Gil

difficulties; and (2) to explore the similarities and
differences between teachers' diagnostic practices in
experimental and classroom situations.

Ten teachers from the Chicago and the Lansing areas
participated in the study. Research was carried out in two
phases. In the first, the teachers interacted with
simulated cases of reading difficulties in a laboratory
setting; in the second, they were observed and interviewed in
their own classrooms.

Each of the ten teachers interacted with two
simulated cases (at two different times). The cases were
built on materials that would be available to the teacher in
a regular classroom setting and made use of children's
natural language patterns. After each interaction, the
teachers went through a stimulus recall session.

A few weeks after the laboratory experiment, each
teacher was observed in her own classroom during one reading
session and interviewed about her reading instruction and
diagnostic practices.

Data analysis consisted of product measures, process
measures, and analysis of classroom interviews. Product
measures dealt with the outcomes of the clinical encounter
between a teacher and a case (e.g., what final diagnostic
judgments were made about the case; what cues were collected,
etc.). Process measures focused on the manner in which the

problem was diagnosed (e.g., length of interaction, number

Doron Gil
of cues collected, number of final stated diagnostic
judgments, and the relationships between cues collected
and teachers' verbatim comments about these cues).

Classroom interviews were analyzed in terms of (l)
diagnostic categories mentioned in the classroom, (2)
remedial techniques offered, and (3) teachers' responses
to various questions in the interview. Comparisons were
also made between teachers' classroom diagnoses and their
laboratory diagnoses.

The major findings of this study were that: (1)
teachers showed little agreement on their final stated
diagnostic judgments for the cases of reading difficulties,
(2) they apparently lack comprehensive or systematic
approaches to gathering and evaluating information about
cases of reading difficulties, and (3) they use the same
global diagnostic categories in laboratory situations and

in the natural classroom environment.

ACKNOWLEDGMENTS

I wish to express my sincere appreciation to those
who have helped me throughout this investigation:

Dr. John Vinsonhaler, who directed my dissertation,
for his on-going help, feedback, and trust in my ability to
conclude this complex study.

Dr. Donald Freeman, chairman of my committee, for
his everlasting encouragement and critical comments on
previous drafts of the dissertation.

Dr. Lee Shulman, for introducing me, together with
John, to this fascinating area of clinical decision making,
and for helping me reach decisions regarding the objectives,
form and methods of this study.

Dr. George Sherman, for teaching me reading diagnosis
on an informal basis, and devoting many hours helping me with
this study.

Dr. Robert Hatfield, who was always helpful with his
constructive feedback on previous drafts of the dissertation.

Dr. Rebecca Barr, from the University of Chicago,
for developing the simulated cases used in this study,
consulting with me regarding the study's methods of data
collection, and enabling me to use the reading clinic
at University of Chicago for part of my observations.

Mrs. Annette Weinshank, who volunteered her time
to assist me with data collection in Chicago.

Mr. Chris Wagner and Ms. Francine Kitchen, who
designed the computer programs needed for data analysis,
and helped me with those analyses.

Mrs. Ruth Polin, Ms. Debie Salters, Mr. Ray Smith
and Mrs. Jerri Zemper, for providing me with the technical
and administrative assistance so vital for this enterprise.

My wife, Rakel, for her continuous spiritual support.

ii

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

Chapter

I.

II.

Introduction . . . . . . .
The Problem . . . . . . .
Research Tasks . . .
Significance of the Study
Rationale for the Study. .
Overview of the Study . . . .

THE PROBLEM O I O O O O C O O C C O O O 0

REVIEW OF THE LITERATURE . . . . . . . . .

IntIOduction O O O O O O I O O O O O O O 0

Nature of Reading Diagnosis . . . . . .

Place of diagnosis in the teaching-
learning situation . . . . . .
Importance of reading diagnosis .
Efficacy of diagnosis . . . . . .
Ways of defining reading diagnosis.
Levels of reading diagnosis . . . .
Nature of reading diagnosis: Summary
Competence of classroom teachers as
diagnosticians . . . . . . . . .
Research on diagnostic skills of
classroom teachers . . . . . . . . .
Models of reading diagnosis . . . . . .
Barr's Model of diagnosis . . . . . . .
Sherman's Model of Reading and Learning
to read . . . . . . . . . . . . . . .
The literature on reading diagnosis:
Conclusions . . . . . . . . . . .

Clinical Studies of Reading Diagnosis at

the Institute for Research on Teaching

Introduction . . . . . . .
The Inquiry Theory . . . .
The agreement corollary . .
Case simulation . . . . . .
Observational studies . . .
Computer simulation studies

iii

Page

add-bu)!“ l-I

£000

13
23

33
35

40
47
47

49
52

54
54
55
57
62
63
65

 

 

Chapter

Training studies . . . . . . . . . . .
Clinical studies of reading diagnosis:
Conclusions . . . . . . . . . . . . .

Summary

III. DESIGN AND METHODOLOGY . . . . . . . . . .

Introduction.
The Subjects

Design .

Phase 1: Laboratory conditions . . . .
Phase 2: Classroom interviews . . . . .
Data Analysis . . . . . . . . . . . . .

Comparisons between teachers' diagnostic
processes in laboratory and classroom
situations . . . . . . . . . . . . . .

Summary .

IV. RESULTS .

Introduction .
Presentation of Results . . . .

Product measures .
Process measures .

Part 1.
Part 2.

Part 3:

Summary .

Classroom interviews
other findings 0 O O O O O O O O O

O C O O O O O O O O O O O O O

V. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS .

Introduction . . . . . . . . . . . . . . .
Procedures . . . . . . . . . . . . . . .
Data Analysis . . . . . . . . . . . . . . .
Results . . . . . . . . . . . . . . . . .
Conclusions . . . . . . . . . . . . . . .
Recommendations . . . . . . . . . . . . . .
APPENDICES
APPENDIX A: CASE INVENTORY . . . . . . . .
APPENDIX B: DIRECTIONS FOR OBSERVATIONAL
SESSION . . . . . . .
APPENDIX C: CLASSROOM INTERVIEW . . .
APPENDIX D: DESCRIPTIONS OF CHILDREN WITH
READING DIFFICULTY . . . . .
APPENDIX : FREQUENCIES OF DESCRIPTIONS OF
READING DIFFICULTIES SELECTED
BY TEACHERS IN THE CLASSROOM .
APPENDIX F: DIAGNOSTIC DOMAIN CASE 7 AND 8

iv

Page
67

69
69

72

72
72
74
78
81
83

160
160
167

169

170
180

184

193
194

APPENDIX G:

APPENDIX H:

APPENDIX I:

LIST OF REFERENCES

LENGTH OF INTERACTION, NUMBER OF
CUES COLLECTED AND NUMBER OF
DIAGNOSTIC JUDGMENTS FOR CASES

7 AND 8 AND 10 TEACHERS
DIAGNOSTIC DOMAIN BASED ON
CLASSROOM INTERVIEWS
OBSERVATIONAL STUDY DATA

ANALYSIS SYSTEM (OSDAS)

STATISTICS

Page

199
200

203

208

LIST OF TABLES

Table Page

1 A comprehensive model of reading diagnosis . . . 34

2 Teachers, school and grades background . . . . . 75
3 Overall design of the study . . . . . . . . . . 77
4 Assignment of cases and inventor forms. . . . . 82

5 Most frequently mentioned diagnostic categories
for Case 7 and Case 8 and 10 teachers . . . . 98

6 Most frequently collected cues for Cases 7 and 8
and 10 teachers . . . . . . . . . . . . . . . 101

7 Diagnostic commonality score for Cases 7 and 8
and 10 teachers . . . . . . . . . . . . . . . 106

8 Cue commonality scores for Cases 7 and 8 and
10 teachers . . . . . . . . . . . . . . . . . 107

9 Inter-clinician correlations for diagnostic
judgments for Cases 7 and 8 and 10 teachers . 109

10 Inter-clinician correlations for cues collected
for Cases 7 and 8 and 10 teachers . . . . . . 110

11 Length of interaction for Cases 7 and 8 and 10
teaChers O O O O O O O O O O O O O O O O O O 116

12 Number of cues collected on Cases 7 and 8 and
10 teachers . . . . . . . . . . . . . . . . . 117

13 Number of diagnostic judgments for Cases 7 and
8 made by 10 teachers . . . . . . . . . . . . 119

14 Correlations among length of interaction, number
of cues collected and number of diagnostic
judgments for Cases 7 and 8 and 10 teachers . 120

vi

Table

15

16

17

18

19

20

21

Number of verbatim comments, mean cue percent
time and verbatim percent time for Cases 7
and 8 and 5 teachers . . . . . . . . . . . .

Twenty-eight most frequently mentioned
diagnostic categories in the classroom
(Total of 96) . . . . . . . . . . . . . . .

Number of diagnostic categories mentioned by
teachers in laboratory and classroom
Situations O O O O O O O O O O O O O O O O 0

Most frequently mentioned diagnostic categories
in laboratory and classroom settings . . . .

Most frequently mentioned diagnostic categories
for 1977 observational study (8 clinicians,
4 cases)* . . . . . . . . . . . . . . . . .

Most frequently mentioned diagnostic categories
for Cases 7 and 8 and 10 teachers* . . . . .

Diagnostic and commonality scores for the
classroom teachers study and the 1977 study.

vii

Page

125

131

136

139

163

164

165

LIST OF FIGURES

Figure Page

1 Percent of diagnostic categories for
Cases 7 and 8 and 10 teachers . . . . . . . 100

2 Percent of all cues collected for Cases 7
and 8 by 10 teachers . . . . . . . . . . . 104

3 Cue to statements relationships . . . . . . . 127

4 Percent of the diagnostic categories mentioned
by teachers in the classroom . . . . . . . 133

viii

CHAPTER I

THE PROBLEM

Introduction

 

Many of today's students have reading deficiencies,
a problem which is of major concern among educators (Satz,
1977). Many children, taught under a skill-model, seem to
have failed to acquire some basic reading technique(s) in
the early grades: Among those are word identification
skills, phonics and structural analysis, comprehension, etc.
The sources of the problems, educators claim, are multiple.

Because classroom reading instruction is deve10p-
mental (Austin, 1968; Carter & McGinnis, 1970, p. 28),
reading deficiencies, if uncorrected, may become pro-
gressively worse, and might interfere with general learning.
Wilson (1977) maintains that "inability to read coupled with
the lack of desire to read, leads directly to school failure"
(p. 1). Thus, if the school teaches the child anything, it
is what failure means (Smith, Carter & Dapper, 1970, p. 4).
Furthermore, the fear of failure becomes more intense as the
child grows older (Reich, 1962) and this, in turn, might
result in a number of emotional problems: tension, anxiety,
frustration, short attention span, and inability to do
independent work (Smith, Carter & Dapper, 1970). These

1

2
problems may impair personal development and behavior
(Whitcombe, 1976; Medrano, 1977). Thus, children with
reading difficulties do not perform at their potential level
of effectiveness, since every phase of their academic and
personal development is directly or indirectly affected by
their reading problems (Sherman, 1968).

Classroom teachers must be able to identify reading
difficulties early and to diagnose and treat them. Early
identification and correction may prevent problems from
becoming more complex and, hence, more difficult to treat
(Bond & Tinker, 1967). Early treatment might also prevent
many of the learning and emotional problems that accompany
reading deficiencies.

Unfortunately, a number of reading consultants
believe that, at present, the retarded reader is not
identified and treated or referred to help early enough
(Carter & McGinnis, 1970). Classroom teachers, some
researchers feel, seem to lack the basic clinical skills
necessary to engage in these preventive processes (Adams,
1971; Stephens, 1978). This is a serious problem, because,
according to some educators, "educational adjustment alone
cannot correct a problem which has been ignored for a long
time to which there is a severe emotional problem" (Wilson,
1977, p. 65).

If diagnosis and remediation of reading difficulties
are fundamental aspects of reading instruction, there is a
need to improve the diagnostic and remediation skills of

classroom teachers. For this to be done, one must first

3
study the diagnostic and remediation skills which teachers

currently have.

The Problem

 

The purpose of this study was to investigate the
clinical problem solving behavior of classroom teachers as
they diagnosed children's reading problems, in both experi-
mental and classroom situations. In other words, it was
intended to examine how teachers gather data about children
with reading difficulties and how they diagnose those
difficulties.

In a recent paper about instructional practices in
reading,Pearson and Kamil (1978) argue that the techniques
teachers use for instruction are determined largely by the
models of reading they follow. That is, a teacher might
emphasize print translation or context and memory, or both,
depending on the model he/she‘prefers. The model, or theo-
retical conceptualizations of the reading process, further
influences which sub-reading skills, -- such as decoding,
getting meaning from the printed page, etc., -- are stressed.
Models also suggest certain diagnostic strategies. For
example, those models which "teach to weaknesses" and empha-
size sub-skills, sequencing and automaticity suggest the
following diagnostic steps:

"(1) define a set of objectives, one for each
skill or interest, (2) place them in a logical
sequence, (3) write a test for each objective, (4)
administer the test to a group, (5) examine skill
profiles, looking for peaks and valleys, (6)
beginning with the earliest skill in the sequence,

remediate all the weaknesses" (Pearson & Kamil,
1978, p. 20)

4
Those models which focus on meaning and "teaching to
strengths", on the other hand, suggest these diagnostic steps:
"(1) determine the components of the reading
process in which a student demonstrates proficiency,
and (2) capitalizing on those strengths, help the
student learn how to generate meaning, hypotheses,
gathering whatever data he needs to confirm or dis—
confirm those hypotheses: (Ibid. p. 20)

An integral part of a diagnosis, regardless of the
strategies used, is the teacher's cognitive and diagnostic
processes; e.g., how he/she processes the information he/she
has collected to reach diagnostic and remedial decisions
about the child's problems. As the review of the Literature
section will show, these cognitive processes, collectively
referred to as "clinical problem solving behavior”, have not
yet been researched; hence there is a clear need for studies
such as this one.

This study builds upon research in progress of the
Institute for Research on Teaching (IRT) at Michigan State
University. The main thrust of this research, conducted by
the Clinical Studies component of the IRT, is the
investigation of clinical problem solving behavior in
reading diagnosis (Gil gt_al. 1979). Most research efforts
to date have concentrated on the clinical problem solving
behavior of reading specialists. The present study was

intended to broaden the scope of this research by exploring

the clinical problem solving behavior of classroom teachers.

Research Tasks

Investigation of the clinical problem solving behavior

followed by teachers in diagnosing reading difficulties have

5
rarely been conducted. For that reason, this study was
intended to be, and should be viewed as, exploratory; it did
not attempt to pose and test hypotheses, but rather to
investigate, probe and describe the diagnostic processes
currently employed by teachers.

Theypurposes of the study:

 

The first purpose of the study was to investigate the
clinical problem solving behavior of classroom teachers as
they diagnosed children's reading problems. Toward this end,
10 teachers were asked to interact with two simulated cases
of reading difficulties in a laboratory setting. An
observational session was held, during which the teachers
were instructed to gather information about a case and to
verbalize their thoughts as they attempted to reach diag—
nostic decisions about the case's problems. During a
debriefing session that followed, techniques of stimulus
recall were used to stimulate the teachers' memory of how
they attempted to diagnose the case. Among the research
questions posed were: I
1. What are the diagnostic judgments most frequently
mentioned by teachers about a given case?

2. To what degree do teachers arrive at similar diagnostic
judgments about a child's reading problems?

3. Do teachers spend similar amounts of time diagnosing a
case of reading difficulty?

4. What are the most frequently collected cues about a

given case?

6

5. What kinds of relationships exist between cues teachers

collect and the teachers' verbatim comments about these

cues and the case?

These questions (particularly No. 1,2, and 4) are
based, in part, on an Agreement Corollary derived from a
theory of clinical problem solving behavior named the
"Inquiry Theory"(Vinsonhaler, Wagner & Elstein, 1977).
The agreement corollary states that, the greater the simi-
larity of teachers' "memories", the greater the agreement of
their diagnoses. For a complete discussion of the Inquiry
Theory and the agreement corollary see Chapter II, pp. 55-62.
A second purpose of the study was to explore the

similarities between teachers' diagnostic practices in
experimental situations and their practices in the natural
classroom environments. Interviews were conducted with the
10 teachers in their own classrooms. The interviews were
designed to provide information about the teachers'
instructional, diagnostic, and remedial practices in the
classroom. Using a standardized interview-form, the
researcher asked the teachers to match their students with
descriptions of reading difficulties (one student to one
description), and to describe how they had determined the
students' reading problems and what remedial actions they
had taken. Specific research questions included:
1. What are the diagnostic categories most frequently

mentioned by teachers in the classroom?
2. To what degree do teachers mention similar or different

diagnostic categories in laboratory and classroom

situations?

7
3. To what degree do teachers formulate similar or different
remediation plans across two cases or reading diffi-

culties?

Significance of the Study

 

Little is known about the effectiveness of class-
room teachers in diagnosing and remediating reading
difficulties. A review of the literature reveals a lack of
empirical research in this area and an absence of descriptions
of the teachers' diagnostic processes.

The data from this study should add to our under-
standing of the clinical problem solving behavior of
clinicians. The study should also help researchers plan and
execute further research on clinicians' cognitive processes.
And, it should help teacher educators seeking methods for
training inservice and preservice teachers in diagnostic and

remediation skills.

Rationale for the Study

 

Many of today's students have reading deficiencies
which may lead to severe learning and emotional problems.
Early identification is essential to remediate these
problems and to prevent them from becoming more serious.
Because reading specialists usually treat only the more
severe cases of reading difficulties, the task of early
identification must be given to classroom teachers, who
interact with and observe the students' reading behavior
on a continuous basis. This study should serve as a first
step toward determining how teachers might be trained for

this role.

Overview of the Study

A review of the literature pertinent to this study is
presented in Chapter II. The review focuses on two areas:
(1) the nature of reading diagnosis and its implications
for classroom teachers, and (2) recent studies of clinical
problem solving behavior in reading diagnosis upon which
the present study is drawn.

Chapter III is devoted to a discussion of the
research design and methods of the study. Data, treatment
of data, and results are presented in Chapter IV.

A summary, conclusions, applications for further
research, and implications for teacher education are

provided in Chapter V.

CHAPTER II

REVIEW OF THE LITERATURE

Introduction

 

The purpose of this chapter is to review the
current literature on reading diagnosis. Two major
areas will be discussed: 1) the nature of reading
diagnosis and its implications for classroom teaching,
and 2) recent studies of clinical problem solving behavior
in reading diagnosis conducted by the Clinical Studies
Component of the Institute for Research on Teaching at
Michigan State University upon which the present study

is drawn.

Nature of Reading Diagnosis

 

Place of diagnosis in the teaching-learning situation

 

Although there is an abundance of books and papers
about reading instruction, not a great deal is yet known
about reading problems, their causes and remediation
(Smith, Carter & Dapper, 1970, p. l). Nor, it seems, do
researchers know enough about the developed skills of the
fluent reader (Smith, 1971). The general notion, however,
is that reading is a more complex process than was

originally believed (Sawyer, 1974).

9

10

Many educators also seem to feel that, since
reading is a complex process, the teaching of reading,
reading diagnosis and remediation are all parts of one
interrelated process. The section that follows con-
centrates on the diagnostic aspect of reading; specific
attention will be given to the importance of diagnosis,
definitions of the process, and its effectiveness, as

suggested by the literature.

Importance of reading diagnosis

 

Review of the literature reveals that most educators
consider the diagnosis of reading performance an essential
and integral part of reading instruction, and a basis of
all efficient teaching (Otto, McMenemy & Smith, 1973).
Moreover, because the reading process appears to be com-
plicated and reading failure may be attributed to many
compounding factors (Satz, 1977), some educators acknowledge
the need for an interdisciplinary approach to diagnosis (and
to research on reading in general) (Weintraub & Farr, 1976).
This interdisciplinary approach, they argue, should focus
attention on the individual child and his or her learning
problems, and should be carried out by professionals from
various fields (e.g. psychiatrists, psychologists) and
educators who occupy various positions within the school
system (administrators, principals, reading consultants,
and classroom teachers) (Kress, 1965: Smith, Carter &
Dapper, 1970; Hollingsworth, 1970; C.W. Peters, 1977). The

diagnostic process, some maintain, should start in the

11
classroom, because "diagnosis is an essential aspect of
teaching and a preliminary step to sound instruction"
(Sheldon 1968). Therefore, it is argued that teachers be
"armed with the diagnostic and remedial techniques
necessary to instruct students as effectively as possible"
(Wilson, 1977, p. 1). Thus, the teacher is viewed not only
as reading-instructor, but also as diagnostician, who con-
stantly studies the reading strengths and weaknesses of
students in order to improve instruction (Bond & Tinker,
1967).

Why do so many educators view the diagnostic act in
reading instruction as so important? Furthermore, why do
they place such heavy emphasis on classroom diagnosis
(rather than leave the diagnosis to the reading specialist)?
Review of the literature suggests several reasons:

First, there is an increasing number of students
with reading difficulties, on the one hand, and a small
number of reading specialists, on the other, making it
impossible to refer all children with difficulties to
specialists. Therefore, it is important that classroom
teachers develop the necessary diagnostic and remediation
skills to detect and help children with reading difficulties,
and thus reduce their number (Gallant, 1970); specialists
will be able to handle the number of students referred to
them only if the classroom teacher assumes major

responsibilities in diagnosis (Wilson, 1977).

12

Second, it is essential that reading difficulties
be detected and corrected as early as possible (Sheldon,
1965), so that they will not become more complex (Bond &
Tinker, 1967: McCarthy, 1971; Rabinovitch, 1965). The
basic assumption here is that "the earlier the problems
are discovered, the more hope there is for conquering them"
(Smith, 1969, p. 15).

Third, there is a notion that reading is an important
enough skill that every phase of the academic career of the
poor reader, and his adjustment to his environment and
peer-group is directly or indirectly affected by defective
reading ability (Sherman, 1968; Stevens, 1971); if this is
so, diagnosis and remediation of reading difficulties can
enhance a child's concept of himself.

Fourth, reading diagnosis should help teachers plan
and modify instruction to meet the needs of individual
students (Austin, 1965: Bond & Tinker, 1967: Dauzat, 1977;
Dietrich, 1972: Farr, 1971: Karlsen, 1976; Olson & Dillner,
1976; Sawyer, 1968; Swalm, 1973).

The assumption underlying the arguments for classroom
reading diagnosis (rather than clinical diagnosis) is that
since classroom teachers have relatively long and extensive
acquaintances with their students, they are in an Optimal
position to diagnose the students' reading performance

through observations and informal tests (Wilson, 1977).

13

Efficacy of diagnosis

 

Although the prevailing assumption throughout the
literature is that reading diagnosis is an essential part of
instruction and remediation (e.g., "individualized diagnosis
leading to prescriptive treatment is a widely accepted and
acclaimed approach to reading disorders" [Bateman, 1971]),
the literature largely ignores two important considerations:
first, whether early and accurate diagnosis actually promotes
effective remediation, and second, whether the remediation
does, in fact, help the retarded reader. If it does not,
one must ask, of what value is the diagnosis?

Spache (1976) maintains that diagnosis is not always
relevant to remediation, and attributes this problem to two
conditions: (1) "the presence of biased or prejudiced
thinking resulting in limited diagnostic efforts or in
stereotyped remedial programs," and (2) "the differing
emphases in the training of clinicians, who are apt to do
the diagnosis, and of remedial or classroom teachers, who
are prominent in remedial work" (p. 436). In other words,
Space claims that the procedures used in remediation
efforts are not directly related to the detailed diagnostic
findings. In many cases, he says, it appears that the two
processes are carried out by different personnel who do not
communicate sufficiently with each other. He recommends
that treatment procedures be more closely related to

diagnostic findings.

14

Theoretical work dealing with the associations
between diagnosis and remediaton is not the only thing
missing from the literature; there is also an obvious
lack of empirical research in this area. Bateman (1971)
states that a detailed diagnosis of a student's reading
performance should yield data on what she/he needs to be
taught, and how she/he should be taught. "The question
of whether diagnosis makes a difference," Bateman adds,
"must be rephrased in terms of which diagnostic data make
a difference for what purpose." She notes that there are
conflicting Opinions about the efficacy of diagnosis and
little conclusive research to clarify this issue.

Discussing the value of diagnosis to the remedial
act, Della-Piana (1968) cites three studies, conducted over
a period of 33 years, which provide data about the
relationship between diagnosis and remediation. The first
study, by Monroe (1932), dealt with procedures for
systematic diagnosis leading to specific remedial acts.
Thorough diagnoses were conducted with children who
exhibited specific reading deficiencies. Taking into con-
sideration the students' age and I.Q., Monroe compared their
reading achievement with their skills in other areas, such
as math and spelling. Educational profiles were created
for each child which assisted the researcher “in understanding
the child's problem of maladjustment by indicating the
magnitude of the discrepancy or lack of harmony between his

accomplishments" (p. 33). Results showed that remediation

15

based on such an extensive diagnosis yielded significantly
greater gains than ordinary school instruction. The
direct associations, however, between the thorough diagnosis
and the remedial gains were questioned by Monroe herself.
The educational profile, she said, "does not help directly
in understanding the causative factors involved in the
defect,...it does not assist directly in the selection of
remedial methods, although it indicates in some cases
discrepancies between the reading skills which offer a
point of attack. The reading index sets a goal toward which
to work in applying remedial instruction and gives a
satisfactory means of measuring the improvement brought
about by the corrective work" (p. 33).

The second study, by Robinson (1946), demonstrated
that thorough diagnosis, followed by remediation, can
help many "unteachable" children learn to read. Diagnoses
were conducted independently by professionals from various
disciplines -- social workers, psychologists, pediatricians,
neurologists, etc. These specialists then met to evaluate
and plan for remediation. Reading deficiencies had been
caused by multiple factors, Robinson found, and the thorough
diagnoses helped many of the children learn to read. Although
the diagnostic process in this study appeared to be highly
valuable, the question remained "whether similar gains
could be obtained by small-group teaching, without expensive

diagnosis" (Della-Piana, 1968).

16

A third study, conducted by Glad (1965), provided
additional data about the value of diagnosis to subsequent
remediation. Results indicated that students who went
through intensive diagnosis showed an average gain of 1.4
on an oral reading test, compared with a preremedial yearly
average gain of 0.5.

Taken together, these three studies still "leave
open the question as to exactly what role diagnosis plays
in obtaining reading gains" (Della-Piana, 1968, p. 2). In
each study it was unclear whether the thorough diagnosis
was the main determinant which effected the high gains.

Many writers believe that diagnosis is necessary
to avoid "standardized remediation." The diagnosis, they
say, should be oriented toward the student's specific
reading problems. However, recent review of the literature
on the association between diagnosis and remediation, from
1900 to 1979 (Weinshank, in preparation), reveals that,
in general, diagnosis and remediation are far from
interrelate and integrate. Some empirical data provided
by Weinshank supports this claim. A study examining the
relationships between diagnostic decisions made by reading
clinicians and their remedial plans concluded that the
diagnoses and remediations were only modestly correlated
(Weinshank, 1979).

In summary, there is an apparent lack of theoretical
discussion and empirical research on the question of whether

thorough and accurate diagnosis actually determines the

17
effectiveness of the remediation. It seems logical to
believe that the stronger the link between diagnosis and
effective remediaton, the higher the value that can be
attached to the diagnostic act.

To enhance the likelihood that diagnosis will
promote successful remediation, at least three conditions
should be met. First, the diagnosis must be reliable. A
reading specialist working with a particular case of reading
difficulties should be able to reach similar diagnostic
judgments at two different times. One must be aware, how-
ever, that even if a reading specialist is consistent in
his/her diagnoses, it may be that he/she has consistently
reached the same faulty conclusions. For that reason, it
is preferable that at least two specialists independently
diagnose the same case. If both arrive at similar
diagnostic judgments, it is likely that their diagnoses
are reliable and accurate.

A second condition under which diagnosis is important
is that both the diagnosis and the remediation be specific.
It is not enough for a diagnosis to be reliable, it should
also indicate the specific areas in which the child
experiences reading difficulties. Specific treatments
apprOpriate for the student and his/her cognitive abilities
must then be prescribed, rather than some stereotyped
treatments recommended for all children and all types of
problems.

Finally, the remediation must be relevant and

efficacious: that is, it should be directed toward those

18
specific reading problems identified during the diagnostic
act as the cause of the child's reading difficulties.
Remediation will be effective to the degree that it is
relevant.

Suppose, for example, that the diagnosis reveals
that retentive comprehension, one of the four vital signs,
is the problemrarea for a particular child [The four vital
signs are: (l) instant word recognition; (2) decoded word
recognition; (3) fluent message segmentation, and (4) re-
tentive comprehension (Sherman, in preparation)]. The
remediation would be relevant only if it treated the child's
comprehension. The remediation would be irrelevant, and
ineffective (although it might still be specific), if it
called only for phonic work to treat the problem.

The issue of relevancy incorporates another issue,
that of time. How long must the remediation be continued
for the reading problem to be cured? If the remediation
lasts for an extended period, it might not remain relevant,
because the child may begin to fall behind in another area
of reading. If, on the other hand, the remediation does
not last long enough, the problem may remain uncured, and
might eventually become as severe as it originally had been.

If the conditions of reliability, specificity, and
relevancy are met, we can argue that reading diagnosis of
reading difficulties is indeed valuable, since it is likely

to lead to treatment and cure.

19

The value of reading diagnosis may also be
associated with the success or failure of the remedial act.
Even when the conditions of reliability, specificity and
relevancy exist, if reading remediation does not improve
students' reading performance, reading diagnosis may be of
no value. If, on the other hand, remediation does show to
have effects on students' reading performance, then reading
diagnosis, leading to a differentiated (rather than
stereotyped)remediation, is important.

The question of whether remedial programs result
in long-term or permanent gains for the students is
discussed by Spache (1976). He claims that there are
several flaws in the traditional ways of measuring
students' remedial gains; specifically, he says:

(1) different reading tests are not comparable, so dual
testing yields incomparable data; (2) different reading
tests measure different aspects of the reading performance,
so inter-tests comparisons are relatively meaningless:

(3) measuring the decrease in gap between achievement level
and mental age is also relatively meaningless, because
intelligence and reading achievement are not parallel
attributes: (4) scores from pre and post tests, used to
evaluate short-term remedial reading programs, are faulty
because of the phenomenon of regression to the mean. In
summary, Spache argues, standardized tests that evaluate

remedial outcomes are very narrow and are not reliable.

20

In a subsequent discussion of the factors that

affect the outcomes of remediation, Spache comes to the

following conclusions:

1.

There does not appear to be any specific amount of time
needed for remediation. What is most important is

the number of contact hours with the student, which

may continue even after the treatment has ceased;

The manner in which remediation is conducted --

whether on an individual basis or with a small group
(5-6 students) -- does not really matter, because gains
are related directly to the number of contact hours with
the student and progress is more readily stimulated by
frequent sessions:

The student's initial reading level when remediation
begins is positively related to gains from remediation.
Those students with the highest pre-remediation test
scores tend to achieve the highest post-training scores;
Remedial work with elementary school students can
result in significant, short-term gains, whereas
secondary school students may show greater long-term
profit:

The personality adjustment of the retarded reader is

a significant aspect of his/her problem and may affect
his/her remedial gains. Little is known, however, about
the relationships between personality characteristics

of the retarded reader and permanent reading gain scores.

21

Discussion of the factors affecting remedial outcomes
leads to an important question: What are the long-term
outcomes of remediation? After reviewing 25 follow-up
studies, Spache (1976) concluded that "remedial treatment
apparently does not affect school progress appreciably,
over time" (p. 336). Of 14 studies in which reading tests
were administered as part of a long-term follow-up, 12
indicated no continued growth after treatment, and only two
showed that retarded readers were functioning at levels
normal for their age—grade status. The latter two studies
involved secondary school students and college students,
which is consistent with Spache's conclusion that secondary
students tend to show greater long-term profit than primary
grade students.

Overall, Spache concludes that, in general, follow-up
studies indicate little or no further development of reading
ability as compared to the accelerated rate during the
treatment. One of the studies, for example (Buerger, 1968),
revealed that "pupils who receive remedial reading instruction
demonstrated significant immediate post remedial reading
gains", but "did not make greater long-term educational
progress than a similar control group". Buerger's
conclusion was that "reading disability is, for the most
part, a chronic condition needing long-term treatment.

What is needed after a rather intensive remedial period is
provision for supportive reading assistance during this

follow-up period."

22

Both Buerger and Spache maintain that, although
remedial reading temporarily relieves reading problems,
there is little evidence of long-term gains from single
remedial treatments.

The "chronic condition" mentioned by Buerger, and
resultant need for continuous and supportive reading
assistance, warrant further discussion. They depict a
situation that may be analogous to that in medicine, where
we witness two kinds of problems: chronic and acute.
Whereas acute problems, such as appendicitis, can be
treated and remediated in a single act, many chronic
problems, such as allergy, can not be remediated for years,
or even in a life—time.

It may well be that in reading, we can establish
similar distinctions between chronic and acute problems.
Some problems, such as poor use of phonics in contextual
reading, might be acute, while others may be chronic, such
as poor retentive comprehension resulting from poor
cognitive linguistic structures or low interest in reading.

The possibility that chronic reading problems exist,
coupled with the evidence of minimal long-term effects of
single remedial treatments, emphasizes the need for continued
"health-care" in the area of reading diagnosis and remediation.
This, in turn, suggests several questions about the manner
in which reading problems are currently treated and
remediated: What should the role of the reading specialist
be? Can reading specialists continue to work with children

on an individual basis and for short periods of time only?

23
How skilled should classroom teachers be to conduct reading
diagnosis and remediation on a continuous basis in their
classrooms (rather than referring children for short-term
clinical work in a clinic setting)? To what degree should
the parents be involved in the remedial act?

Further research is needed to determine the
proportion of acute and chronic problems in reading. Such
research, consisting mainly of follow-up studies, should also
investigate some of the questions raised above.

In summary, the literature on reading diagnosis
emphasizes the importance of the diagnostic act for the
retarded reader. It fails, however, to provide theoretical
discussion and empirical data which deal with the association
between diagnosis and remediation and which demonstrate
whether thorough and accurate diagnosis leads to more
effective remediation. The literature is also inconclusive
about the long-term effects of remediation. Much research

is needed to investigate these two issues.

Ways of defining reading diagnosis

 

The terms "diagnosis of reading" and "reading
diagnosis" are defined in a number of ways in the
literature: there does not appear to be a universal
definition. The term "diagnosis" is frequently equated
with "evaluation" and "assessment" (Bamman, 1970; Harris &
Sipay, 1975; Harris & Smith, 1976), where all these terms are

taken to mean: identification and measurement of the

24
students' level of performance in reading. Even when the
term "diagnosis" itself is used, it is frequently interpreted
in different ways.

The literature reveals that some educators view
diagnosis as a process, while others see it as a product.
Furthermore, some are concerned with the student's general
reading status, while others focus primarily on the student's
weaknesses which need remediation.

The definition of diagnosis as a process emphasizes
the sequence of tasks the teacher performs to determine a
child's reading performance. Wilson (1977), for example,
states that "diagnosis implies that the educator will
actively search for clues to assist in evaluating the present
state of the reader's skill development" (p. 19). Spache
(1976) defines diagnosis as a continuous process of
prOposing hypotheses about a child's performance and testing
those hypotheses (p. 9). Guszak (1972) views the process as
a sequence of relatively simple determinations of a pupil's
reading achievement level, his/her achievement potential,
and his/her prominent skill needs. Peters, N.A. (1977)
defines diagnosis as an "interrelated process through which
a teacher attempts to ascertain the specific strengths and
weaknesses of individual students".

Similar definitions are offered by others who view
diagnosis as a process, but who emphasize the child's
weaknesses. For Harris (1972) diagnosis is "nothing more
than the application of a straightforward, common sense,

problem-solving approach to the study of children who have

25
difficulties in reading" (p. 403). The purpose of this
problem solving approach, says Harris, is "to find out what
is wrong, what caused the difficulty, and what can be done
for it". Similarly, Strang (1964) states that "diagnosis
puts more emphasis on defining the nature of the individual's
reading difficulties and the conditions causing them".
Strang notes, however, that diagnosis is also concerned with
those positive factors of the child's reading on which one
can build a program for remediation.

When diagnosis is described as a product -- as the
explanation or description of the child's reading performance
-- the description usually focuses on the child's weaknesses
and needs; that is, diagnosis is seen as a systematic and
rationale explanation of an individual's inability to read
(Carter, 1970; Carter & McGinnis, 1970).

Most of these definitions suggest some
characteristics of a "good" diagnosis. At the very least,
to be effective, diagnosis cannot be a one-time operation.
Rather, it should be a continuous process (Otto et 31.,
1973) which will be combined with instruction and improve
reading through the reinforcement of the child's strengths
and remediation of his/her weaknesses (Strang, 1964). This
continuous process is interwoven with treatment (Smith,
1969: Smith, Carter & Dapper, 1970) and is designed to
evaluate students very carefully before instruction begins,
and on a regular basis (Sheldon, 1968; Wilson, 1977). The

process of diagnosis and remediation should be recycled, as

26
long as the child fails to respond to the remedial
instruction (Bond & Tinker, 1967).

What seems to be missing from the various
definitions of diagnosis are (1) the specific factors
teachers and reading specialists should be looking for
when they try to diagnose a child's reading problems, and
(2) the mental tasks performed by the teacher or the reading
specialist as they determine the child's reading performance.
To say that diagnosis is "problem solving" or a "rationale
explanation" does not explain how teachers or reading
specialists solve the problem -— how they arrive at
apprOpriate judgments and explanations of the child's
reading difficulties. More attention must be given to the
manner in which teachers process information they have

obtained through observations and the use of various tests.

Levels of reading diagnosis

 

As we have seen, the term "diagnosis" or "reading
diagnosis" is not yet defined in a way that enables one to
thoroughly understand the skills involved in the process.
Furthermore, the definitions of this term do not draw a
distinction between two different kinds of diagnosis --
classroom diagnosis and clinic diagnosis. Rather, review
of the literature reveals that educators prefer to talk about
levels of diagnosis, which correspond to reading difficulties
of different severity.

Wilson (1977) describes three types of diagnosis:

informal on-the-spot: classroom, and clinical.

27

Informal on-the-spot is, as the name suggests, an
informal diagnosis of a student's reading status based on
teacher's questions or exercises. It is followed immediately
by adjustment of instruction; this immediacy may prevent
serious reading problems from developing.

Classroom diagnosis is a more formal and planned
process which is carried on by the classroom teacher if the
adjustments made after the informal diagnosis do not solve
the child's problem. The teacher gathers more information
by observing the student as he/she reads and through the use
of a more specific test of skills. After careful evaluation
of this information, the teacher applies the findings to the
instructional situation.

Clinical diagnosis is conducted by a reading
specialist outside the classroom. The specialist takes
precise measures of the student's reading skills and
potential through the use of precise testing and evaluation
instruments.

Smith 33 a1. (1970) list four levels of diagnosis,
identified according to the persons who carry them out:
classroom teacher, reading specialist, psychologist and
clinician.

Classroom teachers usually have the training
necessary to observe their students' strengths and weaknesses.
They can use a group test to gain information about the
students‘ reading performance and, on the basis of that

information, make immediate instructional adjustments to

28
build the strengths and overcome the weaknesses. Classroom
teachers can also make informal observations of the students'
attitudes, willingness to participate in class, articulation
and speaking vocabulary.

Reading specialists can analyze the reading process
more thoroughly. Using individual diagnostic procedures,
they can measure the gap between a student's achievement and
ability and suggest specific treatments to overcome whatever
weaknesses there might be.

Psychologists and clinicians are called in when it is
necessary to go beyond the student's observable strengths and
weaknesses and beyond measures of the child's reading level.
In such cases, mental and physiological tests might be
necessary.

While Wilson and Smith 35 El: differentiate between
classroom diagnosis and clinical diagnosis, most educators
do not make this kind of distinction. Rather, they prefer
to classify diagnosis according to levels of specificity
and, in so doing, they seem to suggest that there is no
clear-cut distinction between classroom and clinic diagnosis.
Bond and Tinker (1967) describe three levels of diagnosis:
general, analytical and case-study. While many children's
reading problems need only a general diagnosis, others, few
in number, may need an analytical or case-study diagnosis.

General diagnosis is designed "to identify children
who are doing relatively poor work in reading as compared

with their other achievements" (p. 155). The purposes of

29
this kind of diagnosis are to gain the information necessary
to adjust instruction to the needs of individual students and
to locate those students whose problems require more detailed
analysis. This information is obtained through the use of
group tests and other sources of information available to the
teacher.

Analytical diagnosis is a systematic exploration of
specific reading strengths and weaknesses. It is intended
to locate specific areas of reading difficulties.

Case-study diagnosis is used to deal with those
students whose reading difficulties have not been adequately
diagnosed by the general and analytical approaches. Under
this approach, the diagnostician examines the student's
reading potential as well as other physiological and
psychological characteristics.

Like Bond and Tinker, Kennedy (1971) has identified
three levels of diagnosis: group, individual, and clinical.

Group diagnosis is intended to identify students who
have reading difficulties and need corrective instruction.

In this approach, students are grouped according to age or

grade level and given a series of mental, achievement, and

diagnostic group tests. This is the most elementary level

of diagnosis and, according to Kennedy, is used extensively
by classroom teachers.

Individual diagnosis is designed for those students
"who are very slow or those suspected of having problems too
difficult to treat in a regular classroom" (p. 102). Here,
students take individual tests administered by reading

specialists.

30

Clinical diagnosis is a very thorough and compre-
hensive process, conducted by educational diagnosticians
with the aid of other specialists, such as physicians,
psychologists, eye and hearing specialists, etc. At this
level, the diagnosis is intended to provide clues to the
causes of the reading problems.

Strang (1968; 1969) and Carter and McGinnis (1970)
have also categorized diagnostic sequences according to
levels of complexity. Strang has identified seven such
levels, beginning with description of reading performance by
the classroom teacher (Level I) and ending with intro-
spective reports (Level VII), in which the reader is asked
to describe the processes he uses in reading selected
articles. The five intermediate levels include finding clues
to different aspects of the student's behavior that influence
his/her reading performance (Level II); analysis of the
student's reading process (Level III): identification of
the student's mental abilities which underlie success in
reading (such as visual memory and association) (Level IV):
analysis of personality traits and values (Level V), and
neurological examinations (Level VI).

Carter and McGinnis (1970) list four levels of

diagnosis: Level I -- identification of the disability,
Level II -- more detailed description of the problem;

Level III -- identification of the individual's reading
needs, and Level IV -- determination of causal factors:

the determination of why the individual does what he does.

31

Otto gt_al. (1973) feel that the first two levels
of diagnosis described by Carter and McGinnis amount to a
single step, and so they suggest that the diagnostic process
consists of three levels: survey, specific and intensive.

Survey diagnosis is carried out by the classroom
teacher and is designed to uncover the strengths and
weaknesses of the class as a whole, and to identify
individual students who seem to have special instructional
needs. This level of diagnosis involves the use of
class-wide testing and evaluation and attempts to evaluate
the success of the developmental teaching program.

Specific level diagnosis is intended to uncover
the reading difficulties of individual students who were
identified at the survey level. This kind of diagnosis is
carried out by either the classroom teacher or a reading
specialist, depending on time, facilities, complexity of the
difficulty, and the classroom teacher's proficiency in the
use of diagnostic techniques.

Intensive level diagnosis is used for those
individuals with the most severe reading difficulties; it is
designed to discover the basic problems which impede learning.
A complete case-study, usually carried on by a reading
specialist, either in the school or in a clinical setting,
is done to gain a general understanding of the student and
his/her needs.

Monroe (1968) divides the diagnostic process into
two types of analysis: descriptive, in which the examiner

tries to describe as completely as possible the nature of the

32
student's reading difficulty, and causative, in which the
examiner tries to understand the factors that have created
this difficulty.

In all of these hierarchical definitions of diagnosis,
it is unclear where, exactly, the line falls between class-
room and clinical diagnosis and where the clinician starts
to diagnose. A careful look reveals that, usually, the
lowest levels of diagnosis are carried on by teachers in a
classroom setting, while the highest levels call for the
services of reading specialists in a clinical setting. What
seems to differentiate classroom from clinical diagnosis,
then, is the thoroughness of the diagnosis and some of the
diagnostic techniques employed. Whereas classroom diagnosis
is based primarily on observations and informal testing
procedures, clinical diagnosis tends to emphasize more
formal and extensive testing (Wilson, 1977). Whereas
classroom diagnosis leads to adjustment of instruction,
clinical diagnosis results in a detailed description of the
child's reading problems and a list of recommendations for
remediation; and, finally, whereas classroom diagnosis is a
continuous process, clinical diagnosis is a one-time
assessment.

Another deficiency of the literature is that it does
not deal with the question of how different environmental
conditions influence reading and reading problems. A child
who experiences difficulties in the classroom setting may
exhibit totally different behavior when sent to a reading

clinic. The problem then becomes one of diagnosing in one

33
context area (clinic) difficulties that are exhibited in and
affected by a totally different context area (classroom).
Related issue which also lacks sufficient discussion, is
that of the remoteness of the remediation from the
diagnosis. Not only can the problems observed in a class-
room setting be different from those exhibited in a clinical
setting: it may well be that the remediation does not
correspond to the diagnosis. This is especially likely
when a teacher plans a remediation on the basis of a
diagnosis made in a clinic (by a reading specialist) or,

even more remote, by a psychologist from outside the school.

Nature of reading diagnosis: summary

 

The literature pertaining to reading diagnosis deals
with three major issues: The purpose of diagnosis (Why),
its content (What) and its methods (How). Table 1
describes this domain of diagnosis in greater detail.

The purpose of reading diagnosis is threefold;
it is (l) necessary for the early identification of
reading problems; (2) a prerequisite for remediation, and
(3) essential for planning, modification, and
individualization of instruction. The content of such a
diagnosis focuses on two areas: skill performance and
causes of reading difficulties and, accordingly, involves
two activities: (1) description of skill performance, and
(2) determination of causality. The methods involve a
continuous process of diagnosis and the employment of one

of three levels of diagnosis corresponding to the severity

34

nomoummm

mumcwamwomwcumucH .oa

Ammmsmo mo
mEHmu cwv Damaommm
mpsum wmm01m>flmcmucH
Hmscw>wccH

scaumucmwuo OHCAHU
Acowumfluomoc mo

mEHmu may owmwommm

HmowuhamCMIHmfiuom

HospH>H©cH

cofiumacwfluo Hoonom

Hmumcoo

HmEuomcH

macaw

COH¥MHCOHHO EOOHm WMHU

mmoooum mooscaucoo

 

mwozwdHQ 30m

coauosnumcw mo
CQADMNHHMSUM>Hpr
can coaumowmapoa
muflHMmsmo mo coflumcfieumuoa .m .mcwccmam How Hmwucmmmm .m

 

 

ucoEm>ownom mam Hofiucmuom cowumacwsmu How
cmo3umn hocmmmuomwc An ouflmwswmuoum mummmoomz .N
mocmEHOMme mEmHnoum mcflpmmu mo
Hawxm mo cowumfiuomoa Am .v coaumowmwucmcw hanmm .H
mHmozw<HQ 4 mmBDBHamzou Emma mmozwdHa um:

mflmocmmap mcflcomu mo H0608 o>Hmcm£mHmEoo «1|.H mamaa

35

of the problem: (a) classroom, (b) school, or (c) clinic.
It may well be that an effective model of reading diagnosis

should incorporate all of these elements.
Competence of classroom teachers as diagnosticians

Something which seems to be missing from the
literature is a discussion of how an effective diagnosis
and remediation should be conducted. Different educators
express different opinions about the purpose, content and
methods of diagnosis, but their writings tend to be very
general and do not describe how information (Table 1)
should be put together to make decisions about individual
students.

Although there appears to be no comprehensive
description of diagnosis -- of how to process information
after is is collected -- many educational authorities
believe that teachers are qualified to diagnose students'
reading difficulties. They consider classroom diagnosis
an integrative part of effective teaching: in fact, some
go so far as to suggest that all content-area teachers
should be involved in reading diagnosis (Dupuis and
Askov, 1977: Niles, 1965). It might be remembered too,
that some of the diagnostic levels described in the
literature are intended to be carried out by the classroom
teacher. The underlying assumption of the above is, that
classroom teachers are, indeed, capable of diagnosing

students' reading difficulties.

36

Smith (1969) and Smith 33 a1. (1970) identify two
major objectives of reading diagnosis conducted by the
classroom teacher: (1) determination of the appropriate
instructional level for a given student, and (2) deter-
mination of the specific skills the student lacks. The
more capable the classroom teacher is of accomplishing
these objectives, the smaller the number of students will
be who need remedial reading instruction. However, as
Smith (1969) says, "part of the classroom teacher's job is
to know when a disabled reader needs help beyond that
offered in the classroom" (p. 20).

McGinnis (1970) believes that the classroom teacher
"can meet the reading needs of 92 percent of the school
population. (The teacher) can observe the child in the
classroom environment, can identify his problems, and can
help him solve them."

Harris (1972) states that "children who have slight
to moderate reading disabilities are generally able to be
helped considerably by the classroom teacher" (p. 400). To
be a diagnostic teacher, Harris (1977) claims, the teacher
must (1) know the scope and sequence of the total reading
program: (2) be familiar with the materials which foster
independent learning; (3) have the support of administrators
or supervisors, and (4) be skilled in analyzing pupil needs,
organizing the class for differentiated instruction, keeping
records, and motivating pupils. Basically, Harris (1974)

argues, the reading teacher can be a diagnostician "if the

37

term diagnostician is defined apprOpriately, and if the

teacher has the knowledge, the facilities and the skills

necessary to make diagnostic teaching work."

Ekwall (1976, 1977), who also feels teachers are
capable of diagnosing reading difficulties in the classroom,
offers some diagnostic guidelines for teachers who work with
disabled readers (at any given grade level). These guide-
lines are:

1. Gather that amount of initial diagnostic information
which falls somewhere between the position that it
is good to do a great deal of diagnosis before
remediation begins, and the contrast position that it
is better to do only enough initial diagnosis to
initiate flexible remediation;

2. When deciding upon tests to administer to children,
consider each student in terms of the type of problem
he/she exhibits and select the tests which correspond
to this problem;

3. Administer the tests in a situation that is similar
to the actual reading situation.

4. Do not attempt to diagnose those factors for which
it is not possible to provide remediation or for which
remediation has not proven effective in the past, and

5. Diagnosis for a seriously disabled reader should
involve more than an appraisal of educational factors.

Spache (1960) thinks it is possible to conduct

clinical diagnosis in the classroom. He defines the

38
retarded reader as an individual who (1) is behind in a
number Of reading skills by one year or more, if in the
primary grades, or by two years of more if older, (2) is
below the reading level necessary for full participation in
the reading tasks of his/her peer-group; (3) has had normal
Opportunities for schooling, and (4) has continued to show
this degree Of retardation despite corrective efforts over
a period of months. Spache argues that the classroom teacher
can apply this definition to identify those pupils who have
reading problems, and then administer group tests of reading
and intelligence to distinguish severely retarded readers
from those needing moderate corrective treatments; he also
claims that the teacher has the capacity to take further
diagnostic steps with those children needing careful
diagnosis. These steps include assessment of physical
factors, personality, and sociological factors; choice of
teaching approach; and administration Of informal tests.

In a similar vein, Austin (1965) and Bond (1970)
state that classroom teaching should be based on diagnostic
teaching. More specifically, they suggest that teaching
should be based on an understanding of the reading
strengths and needs of each child. This knowledge makes
it possible for teachers to adjust their teaching and to
”make modifications to correct any confusion before it
becomes seriously limiting to a child's future growth in
reading" (Bond, 1970). If such modifications were made,
only the more severely disabled readers would need more

careful diagnostic treatment, which has usually been

39
provided outsine the classroom setting.

Otto, McMenemy and Smith (1973) believe that
teachers are competent enough to conduct diagnoses in the
day-to-day classroom setting and to take the steps required
for effective corrective instruction. Moreover, they argue,
the competent teacher usually knows how far he can go in
diagnosing students' reading problems and when he/she should
refer them to a specialist. Similarly, Kennedy (1971)
suggests that group diagnosis (Level I) which is "conducted
by classroom teachers, can uncover many of the more common
reading deficiencies" (p. 103). He goes so far as to state
that "identifying students who need help in reading is the
concern of classroom teachers; it is not a function Of the
remedial reading staff" (p. 73). Because early identification
and treatment can prevent reading problems from becoming
worse, Kennedy adds, "it is...necessary that each teacher
of reading be prepared to administer a preliminary diagnosis
of their children's reading and be able to present simple
corrective measures which will eliminate many minor and
some major problems" (p. 545).

Smith gt El. (1970), maintain that most classroom
teachers have the necessary training to Observe the strengths
and weaknesses of students in the areas of vocabulary, word
recognition and comprehension. Consequently, these
educators say, teachers occupy the pivotal position for
discovering reading difficulties, treating them, Or
referring the students to reading specialists. Wilson

also claims that the classroom teacher is in the best

40

position to first notice students' potential problem
areas (Wilson, 1977). Hence, as Smith points out, the
more able the classroom teacher is in preliminary diagnosis
and implementing corrective instruction, the fewer the
students who will need remedial instruction. Accordingly,
attempts have been made by practitioners to identify
teaching skills which seem essential for diagnosing
reading difficulties (Blount, 1973) and to disseminate this
knowledge to teachers (Florida Department Of Education, 1977).

As this discussion suggests, for diagnosis to make a
positive contribution to the child, teachers must assume the
responsibility for conducting informal diagnoses in the
classroom setting (Dauzat, 1977; Jan-Tausch, 1971) and
accept the notion that identification and correction of
reading problems is an important aspect Of the teaching job
(Smith gt gt. 1970). It must be remembered, of course,
that a diagnosis made by a classroom teacher is different in
its aims and in its depth from one carried on by a reading
specialist who interacts with a child on a one-to-one
basis (Harris, 1953). For teachers, to carry out their
diagnostic responsibilities, they must have access to rapidly
implemented diagnostic procedures which yield accurate data

(Dauzat, 1977).
Research on diagnostic skills of classroom teachers

Despite the many statments about the teacher's
ability to conduct diagnoses, research on teachers' diag-

nostic skills is still lacking. Acknowledging the need for

41
such research, Burnett (1963, 1970) developed a paper-and-
pencil problem solving test in reading to measure
teachers' skills in using diagnostic procedures. Burnett's
principal task was to construct a measure which would "be
valid in terms Of discriminating levels of proficiency at
problem solving in the teaching of reading, and would as
well hold some promise of providing insight into why
individuals differ in proficiency" (1963). The first step
Burnett took to establish this validity was to develop
problems similar to those encountered in the classroom,
problems whose solutions called for operations similar to
the Operations used by the problem solver in an actual
situation.

The test consisted of sets of problems designed to
measure teachers' proficiency in the use of diagnostic
procedures, at five levels of Operation (1963): (1) selection
Of critical information from a pool of data; (2) selection of
a means for securing additional data; (3) interpretation Of
the data; (4) suggestions on how to improve instruction
based on the information collected, and (5) re-evaluation
of these suggestions in light of new information provided.

At each Of these levels, the examinee was asked
to rank four responses in terms Of how well they met the
specifications called for in the problem. For example, the
examinee may have been asked to rank four explanations Of
a child's reading problem on the basis of how well the

explanations fit the information that had been presented.

42

Three groups were involved in the study: (1) students
enrolled in undergraduate elementary education courses;

(2) elementary-school teachers, and (3) reading teachers,
consultants, and supervisors who met the International
Reading Association's specifications for certification as
reading specialists.

Results indicated that the reading specialists
significantly outscored the teachers at all five levels
of Operation. However, Burnett (1970) noted, "a shocking
finding was that the performance of experienced teachers
in terms of mean scores on the problems was only slightly
higher than that of students in undergraduate courses".

Burnett acknowledged that paper-and-pencil tests
cannot really tell us how effectively teachers might perform
in an actual teaching situation. Another shortcoming of
these tests is that they did not trace the decision making
and thought processes of teachers as they attempt to
diagnose children's reading problems.

A 1971 study (Weule, 1971) of practices used in the
diagnosis of reading difficulties at the primary school
level indicated that teachers' diagnostic skills were
inadequate. In the study, a questionnaire was distributed
to about 350 classroom teachers in New South Wales. The
questionnaire was designed to provide descriptions Of
reading diagnoses made by these teachers; the information
asked for dealt with both stated and actual practice. About

199 of the 350 questionnaires were returned.

43

While asked to consider their own role in diagnosis
and treatment, Weule noted, almost all of the teachers said
they felt it is "the teacher's duty and responsibility to
diagnose and treat reading difficulties. Many underlined
the importance of this task. Many placed the teacher above
everyone else in this job, saying that he is in the best
position to make the decision required...Many stressed the
teacher's role in prevention and early identification".
Moreover, Weule said, "seventy per cent of the sample
thought they could cope with pupils of average ability who
are retarded one year or less in reading achievement.

Forty per cent also felt capable of providing adequate
diagnosis for the seriously retarded, average ability child.
Fifteen per cent felt capable of providing adequate
diagnosis and treatment for disabled readers, and thirty
per cent felt capable of OOping adequately with mentally
retarded readers". However, many of the teachers talked
about "the disparity between what they want to do and what
they could do under existing teaching conditions". Many
felt there was no time available to provide individual
instruction to pupils who needed it.

When asked to rate the relative importance placed
on diagnosis and treatment of reading difficulties and
other aspects of education and teaching, the teachers
indicated that much more emphasis is placed on discipline,

pupil-teacher relationships, daily preparation, and so forth.

44

Weule's conclusions were that "teachers appear to be
rather poorly equipped to carry out a crucial part of their
task as educators. Their relatively poor knowledge of
tests,...the poor use of diagnostic techniques all reflect
seriously upon the teacher's knowledge of the children he is
attempting to teach. At the present moment most diagnostic
attempts appear to be inadequate and ineffectual in providing
for successful treatment of reading difficulties."

More recently, Dupuis and Askov (1977) reported on
a Content Area Reading Project which examined classroom
diagnosis and which yielded more positive results than the
Weule study. The hypotheses examined in the Content Area
Project were: (1) that informal diagnosis of reading levels
is the responsibility of the teacher, and (2) that teachers
can use diagnostic information and teach the necessary
reading skills without neglecting their content respon-
sibilities.

Sixty teachers were instructed to use a Decision
Model for Diagnostic Teaching by Grouping, developed by
Cartwright, Cartwright and Yssledyke (1973), which called
for them to employ a step-by-step planning sequence in
the teaching-learning process. The steps were: (1) careful
assessment of each student to determine his/her reading
ability; (2) specification of teaching goals for each
student; (3) grouping of students according to interests,
needs, and abilities; (4) selection of instructional

strategy and management procedure for each group;

45

(5) selection of instructional materials for each group;
(6) implementation of strategy and materials with each
group; (7) evaluation of each student's performance.

After being instructed on how to use the
instructional techniques for carrying out these steps, the
teachers were asked to indicate whether they felt the
techniques were valuable to them or not. Most of their
evaluations were very positive. Moreover, classroom
Observations indicated that most teachers did, indeed, use
the instructional techniques which they were taught.

There are at least two apparent deficiencies with
the decision model, however. First, even if teachers are
taught to implement instructional techniques, it is unclear
how they are-to make the various decisions necessary for
diagnosis, e.g. how they should go about assessing the
reading ability of each student. In other words, the model
tells them what to do, but not how to do it. Second,
although classroom observations seem to indicate that
teachers do, indeed, use these instructional techniques in
the classroom, it does not indicate how well they use them
in diagnosing students' reading difficulties.

In another recent study Stephens (1978) used a
simulated case of reading difficulty to investigate
teachers' diagnostic skills. The simulated case was based
upon the case history of an eight year-old boy and contained
various diagnostic tests usually administered to individual

children in a clinic setting. The teachers' task was to

46
interact with the simulated case, to gather and examine
information about the child, and to arrive at a diagnostic
judgment about his reading problems.

The study included 30 teachers divided into three
groups of 10 as follows: Group (1) classroom teachers with
training -- experienced elementary school teachers with
graduate level instruction in the diagnosis and correction
of reading difficulties; Group (2) classroom teachers with
no training -- experienced elementary school teachers without
graduate level instruction in the diagnosis and correction
of reading difficulties; and Group (3) Pre-Service
Elementary Education teachers -- pre-service teachers with
no full time teaching experience or graduate level
instruction in the diagnosis and correction of reading
difficulties.

Data analysis included both product and process
measures. The product measures were the final diagnosis
made by the teachers; the process measures were selected
features relating to the manner in which the problems were
diagnosed.

The conclusions were that neither graduage level
instruction in reading diagnosis nor classroom teaching
experience significantly affect teachers' knowledge or
strategies of diagnosis. A possible limitation of the
study, however, was that the teachers were asked to
interact with materials originally developed for the use
of reading specialists interacting with children on a

one-to-one basis in a clinic situation, and not by

47
classroom teachers in an actual classroom setting. This

condition may have had a significant impact on the results.

Models of reading diagnosis

 

If, as the literature on reading diagnosis suggests,
classroom teachers should be diagnosticians, two elements
are needed: (1) models which will more precisely and
accurately describe the aspects of a child's reading
performance and/or the reading process and the factors
which affect reading; and (2) research, based on these
models, which investigates the effectiveness of reading
diagnosis made by classroom teachers.

Presently, in two independent projects, researchers
are working on two such models. One is a Model of Diagnosis,
develOped by Dr. Rebecca Barr at the University of Chicago;
the other is a Model of Reading and Learning to Read
developed by Drs. George Sherman and John Vinsonhaler at
the Institute for Research on Teaching, Michigan State

University.

Barr's Model of Diagnosis

 

The model developed by Dr. Barr is one which
indicates those aspects of a child's functioning level
about which the diagnostician must secure information.
Teachers trained under this model are called upon to
practice diagnoses based on materials which would be
available to them in the classroom for reading instruction

and for diagnosing reading problems. The model makes use

48
of Children's natural language patterns in teaching teachers
to make decisions about reading performance.

The Barr model emphasizes the design and modification
of instructional objectives apprOpriate for the child (Barr
and Sadow, 1978). Training builds upon the teachers' prior
knowledge of reading instruction; it is intended to refine
the teachers' concepts and to help them become more effective
in their instruction. An important assumption of the model
is that teachers do, in fact, possess an ability to process
children's natural language, but that they do not have a
systematic way to process and organize the information
collected to render judgments about reading problems.
Accordingly, this approach is designed to train teachers
to carefully and systematically observe and record
children's natural language in the classroom and to use the
information collected for diagnosis and remediation.

Training emphasizes the assessment of three main
areas of the child's reading develOpment: reading
comprehension, print translation and language concepts. To
make such assessments, the teacher must obtain information
that reveals the children's level of functioning in the
three areas -- oral reading of text or portions of text in
the first case; their responses to comprehension questions
based on printed text, in the second; and responses to
aurally-presented language, in the third. With this
information, the teacher determiens the child's strengths
and weaknesses within each area. For example, in evaluating

a student's proficiency in print translation, the teacher

49
will use oral reading evidence to draw conclusions about
the number of words the child is able to recognize
instantaneously; his/her knowledge of word components,
including letter sound association; and his/her integration
of this knowledge into a systematic strategy for recognizing
print. Similarly, in assessing a student's comprehension,
the teacher will draw conclusions about the child's ability
to respond to questions demanding recall of stated information,
his/her inferential reasoning from stated information, and
his/her application of concepts to new situations.
Comprehension is viewed as being dependent on print
translation skill and familiarity with the verbal concepts
used by the author of the text.

Barr's model defines diagnosis as a process whereby
the teacher identifies the conditions that interfere with
the child's ability to comprehend assigned or selected
materials. Intervention by the teacher may involve:

(l) instruction in print translation, language concepts,
and/or comprehension, and (2) modification of assigned

materials.

Sherman's Model of Readingyand Learning to Read

 

The major assumption on which Sherman's Model of
Reading and Learning to Read (MORAL) is based is that
the process of clinical problem solving is governed by
the interaction of two factors (Sherman, in preparation):

(1) the clinician's theoretic conceptualization of the

50

cleint's potential problems and strengths and of possible

cures, and (2) the clinician's actual experience, which is

a major source of the associations between cues and problems,

and between problems and remediation plans. If this

assumption is correct, then the accuracy and clarity of a

clinician's Model of Reading and Learning to Read determines

the consistency and effectiveness of the clinician's

decisions with regard to diagnosis and remediation.

on

According to Sherman, reading diagnosis is based

the client's performance on four Vital Signs of Reading:

Instant Word Recognition: the ability to recognize
words without hesitation, measured by relative size of
sight vocabulary;

Mediated Word Recognition: the ability to recognize
unfamiliar words through the use of graphemic redundancies;
Phrase Comprehension: the ability to translate groups
of words into cognitive structures with semantic
meaning and to retain these structures briefly, and,
Retentive Comprehension: the ability to retain the
results of perceptive comprehension and to organize
these results into more complex semantic structures
suitable for long-term retention, complex inference,
and problem solving.

Sherman's MORAL holds that the performance level

of a client on most reading activities is a probabilistic

function of performance levels on these four fundamental

reading tasks. Therefore, it seems that one of the

51
components of reading diagnosis should be the description
of a child's performance in these four areas. For many
teachers, and some clinicians, establishing the child's
levels of performance on the four vital signs is the
diagnosis.

Sherman argues that diagnosis must proceed beyond
mere surface indications of vital signs. The diagnostician,
he says, must attempt to associate each vital sign with
specific effectors (causes), such as subskill prerequisite
knowledge and specific learning paradigms (e.g., assumptions
about learning) and with specific causal factors (e.g.,
observable and sometimes manipulatable characteristics
of the client and his/her environment). Sherman's belief
is that each Vital Reading Sign performance level can be
improved if the diagnostician has an understanding of the
different effecting antecedents and conditions of learning.
He assumes that the learning necessary to improve each sign
is modulated by a different (but not exclusive) set of
factors.

The different antecedents of learning for Instant
Word Recognition, for example, might be cue discrimination
and associative strategies. The conditions for learning
might be repetition of recognition (practice), motiviation
and feedback.

The effecting child variables may be inferred from
the Learning Paradigm. They are characteristics of the

client and his/her environment which influence Antecendent

52
Processes and Learning conditions. There are three groups
of such factors: Child (e.g., cognitive and affective
factors); School (e.g., academic length of interaction,
regular training given to students), and Home (e.g., the
student's language environment).

In summary, Sherman's Model of Reading and Learning
to Read implies that effective clinical diagnosis and
remediation is determined by the clinician's theoretic
conceptualization of reading as well as his actual
experience. Diagnosis, according to Sherman, should proceed
beyond mere examination of the four vital signs and
consider the various antecedents and conditions for
learning associated with the child's performance on each of

the vital signs.

The literature on reading diagnosis: conclusions

q

 

The purpose of this section has been to review
the current literature on reading diagnosis with emphasis
placed on that information which is pertinent to the
objectives of this study. Conclusions suggested by this
review are as follows:

1. Most educators seem to agree that reading diagnosis
in the classroom is an integral part of the
teaching-learning process and is a prerequisite to
effective teaching.

2. Educators agree that reading difficulties tend to
hinder students' intellectual and emotional

development. Thus, early and continued classroom

53
diagnosis designed to prevent reading problems from
developing is fundamental.
The term "reading diagnosis" has different
connotations; e.g., process vs. product; state of a
child's reading performance vs. weaknesses. Furthermore,
there appears to be several levels of diagnosis (e.g.,
classroom, school, clinical) used to identify and help
groups and individuals with reading difficulties; the
level used depends upon the severity of the problem.
The line, however, between classroom diagnosis and
clinical diagnosis is not well drawn. Moreover, neither
the definitions of reading diagnosis nor the description
of its levels describe the mental tasks performed by the
diagnostician. Overall, what seems to be missing is
a comprehensive model of reading diagnosis which would
put together all the elements that should be involved
in an effective diagnosis.
An implicit assumption implied throughout the literature
is that classroom teachers are capable of diagnosing
students with reading difficulties. (Some educators
even argue that teachers should be accountable for
such a diagnosis). There is little empirical data,
however, to confirm or disconfirm this notion.

In general, the literature fails to provide an

adequate description of the diagnostic processes employed

by classroom teachers, nor does it provide substantive

empirical evidence of the competence of classroom teachers

in reading diagnosis.

54

The present study attempted to respond to these two
deficiencies by providing comprehensive data about teachers'
diagnostic skills. The data was obtained through
investigation of the problem solving behavior of classroom
teachers as they interacted with cases of reading difficulty.
By using this approach we also sought to describe the mental
tasks involved in the diagnostic process.

This study was one of a series of interrelated
studies of clinical problem solving behavior in reading
diagnosis that are being conducted at the Institute for
Research on Teaching (IRT) at Michigan State University.

A discussion of the general research project will provide
the reader with a better understanding of the present study,
and a framework with which to examine it. This discussion
may also be viewed as an introduction to the third chapter
on design and methodology.

Clinical Studies of Reading Diaggosis at the
Institute for Research on Teaching

Introduction

 

The present study is a part of a series of
investigations of the process of effective reading diagnosis
and remediation which was begun three years ago by the
Clinical Studies Component of the Institute for Research
on Teaching (Gil gt gt., 1979). These studies are based
on a theory of clinical problem solving known as the

"Inquiry Theory" and have two general objectives:

55

1. Better understanding of clinical problem solving
behavior of reading clinicians and classroom teachers;
and

2. Application of research findings to the training of
teachers and reading clinicians in diagnosis and
remediation skills.

Three different forms of studies are being

conducted (Gil, Vinsonhaler and Wagner, 1978):

1. Observational studies, in which reading clinicians or
classroom teachers were Observed as they interact with
simulated cases of children with reading difficulties;

2. Computer simulation studies, in which the implications
of the Inquiry Theory are explored by programming
computers to simulate the diagnostic and remediation
behavior described by this theory. (These "simulated"
clinicians reflect both ideal and typical approaches
of reading clinicians to diagnosis and remediation).

3. Training studies, in which instruction in reading
diagnosis and remediation in classes is guided by
the Inquiry Theory concepts. Students enrolled in
the classes interact with simulated cases of children
with reading difficulties, and are guided by computer-

based decision aids, to improve their diagnostic skills.

The Inquiry Theory

 

The Inquiry Theory is a theory of clinical problem

solving, which originated in the medical education research

56
"Inquiry Project" conducted at Michigan State University
by Arthur Elstein and Lee Shulman (Elstein, Shulman, and
Sprafka, 1978). It was refined by the Clinical Studies
group at the Institute for Research on Teaching and has
come to be known as the "Inquiry Theory" (Vinsonhaler,
Wagner, and Elstein, 1977).

The Inquiry Theory describes the behavioral domain
in which a clinician (e.g., a physician, a reading
specialist, a teacher) interacts with a case (e.g., a
patient, a student) to reach diagnostic and treatment
decisions about the case's problems.

By describing this behavioral domain, the Inquiry
Theory attempts to predict those characteristics of the
clinical interaction which will occur (and be observed)
over and over again.

The theory holds that clinical diagnosis, or
clinical problem solving, is determined by the interaction
of clinical memory (consisting of sets of problems, cues,
treatments and the relationships among them) and clinical
strategy (the sequencing of the mental tasks performed by
the clinician), and the case.

Six information processing tasks characterize the
clinical encounter between a clinician and a case:

1. Cue acquisition -- gathering of data, beginning with
initially available cues.

2. Hypothesis generation -- formulation of alternative
assumptions about the case's problems based on a

limited number of cues.

57

3. Cue interpretation -- interpretation of the data based
on the cues collected and the hypotheses under
consideration.

4. Hypothesis evaluation and diagnositic judgment --
estimation of the likelihood of each hypothesis in
order to reach a diagnostic judgment.

5. Treatment evaluation -- estimation of the expected
gain for each of the possible treatments in relation
to the diagnosis.

6. Prescription selection —- selection of the specific
methods of remediation on the basis of the possible

treatments.

The agreement corollary

 

As stated, the Inquiry Theory holds that clinical
problem solving behavior is a function of the clinician's
memory and strategy. This assumption suggests the following
agreement corollary: If diagnosis is determined by clinical
memory and strategy, then similar memories and strategies
will result in similar diagnoses.

According to the agreement corollary, memory and
strategy are determined, in part, by the training
clinicians receive.. The more similar the clinicians'
training, the more similar their memories and strategies
will be, and the more closely they will agree when
diagnosing the same cases of reading difficulties. To

the extent that their training varies, the clinicians will

58
employ different "memories" or "models" of the reading
process, and will show limited agreement in their
diagnostic judgments.

Suppose, for example, that for reliability reasons,
two teachers, with different training backgrounds, are
asked to independently diagnose the same case of reading
difficulties. Assume that one teacher operates from a
skill-model orientation, the other from a socio-psychological
model. Both teachers will use the same initial informal
reading assessment procedures, such as analysis of school
records and oral reading, to evaluate the case's reading
performance. Their different training background, however,
will affect their subsequent choice of information about
the child's reading performance. The first teacher will
tend to collect information relevant to his/her skill-model
orientation, such as the child's ability to chunk, analyze,
and identify words, while the other will probably seek
socio-psychological information, such as psychological
reports, the social-worker's comments about the child,
and home background. The differences in the information
collected will probably lead the teachers to different
diagnostic judgments about the child's reading problems.

The following is a list of possible diagnoses which
the two teachers might entertain. (In other words, it is
a list of some of the diagnostic categories "stored" in
each teacher's memory, based on, and derived from, the

training the teachers have received):

59

 

 

Skill orientation Socio-psychological orientation
poor phonics* low interest in reading*

poor word analysis poor visual acuity

poor oral reading* reads below grade level*

poor comprehension poor auditory acuity*

poor sight words* inadequate verbal ability

reads below grade level* no intellectual models of

reading at home

Suppose the asterisk (*) indicates those
diagnostic judgments each teacher made about the child,
based directly on the model from which he/she operates.
A careful look at these judgments indicates that, although
each of the teachers might have mentioned some of the child's
problems, their diagnoses cover only a small part of the
child's reading performance. The skill-oriented teacher,
for example, mentioned "poor phonics" as a problem area,
but did not suggest a possible cause for this problem,
which may be "poor auditory acuity" (mentioned by the
second teacher). She/he also failed to identify the cause
of the child's "poor sight words," which may very likely
be "low interest in reading" (mentioned by the second
teacher).

Teachers' diagnostic agreement can be determined
by various statistical procedures, such as calculation of
the proportional agreement statistic, commonality score, or
inter-clinician coefficients (See Appendix I: Observational
studies data analysis systems [OSDAS] statistics). In this

instance, the inter-clinician phi coefficient yields a score

60
of around 0.0, meaning that the two teachers showed almost
no agreement in their diagnoses of the child's problems; of
a total of seven diagnoses mentioned by the teachers, only
one (”reads below grade level") was made by both. This
limited agreement is a result of the different kinds of
training the teachers received. The teachers developed
different "memories" about reading and reading deficiencies,
and they considered only those factors that could be
derived from the narrow list of categories stored in those
memories.

It should also be noted that neither teacher's
diagnosis, considered separately from the diagnosis made
by the other teacher, was sufficiently reliable to form
the basis for remedial plans for the child. A combination
Of their diagnoses would probably be more accurate,
comprehensive, and reliable, and lead to more effective
remediation.

The agreement issue is an important one for all
professions in which clinicians must make diagnostic
decisions about their clients. The importance of the issue
stems from the assumption that the higher the clinicians'
agreement about a patient's problems, the more likely it
is that they have reached accurate and reliable diagnoses
(which, in turn, may lead to apprOpriate treatment).

The question of agreement is especially important
in.medicine, where incorrect diagnosis can result in
permanent damage to the patient, and, in some cases, even

in death. Because of the seriousness of the issue, many

61

studies have been conducted to investigate inter-physician
agreement. Several investigations have revealed that "a
surprising and clinically important degree of inaccuracy is
to be expected in the interpretation or evaluation of many
clinical and laboratory procedures used in every day
practice" (Garland, 1959). For example, writes Garland,
"experienced interpreters of a series of plain chest
roentgenograms are apt to miss about 30 percent of those
films positive for roentgenologic evidence of disease,
and to over-read about 2 percent of those negative for
disease. Further, in evaluating pairs of serial
roentgenograms for alternations in the status of disease,
one experienced physician is apt to disagree with another
in about one-third of the cases, and (on review) to
disagree with himself on one-fifth of them. Comparable
degrees of error occur in many forms of clinical practice.
Indeed, if all branches of medicine could be tested, the
phenomenon would probably be found quite universal".

Studies by other researchers support Garland's
statements. In an examination of the reliability of chest
radiography in the diagnosis of pulmonary lesions,
Yerushalmy (1955) found that an individual physician
agreed with other radiologists on only about 30 percent
of the x-rays. Yerushalmy concluded that "observer error
in chest radiography is of such magnitude as to create
special problems in diagnosis and treatment." In another
study, Paton (1957) found that "the accuracy rate in the

diagnosis of myocaridal infractional is surprisingly low,

62
only 44 percent."

Given the importance placed on diagnostic agreement
in medicine, it seems logical that this is an important
consideration in reading diagnosis, as well. For this
reason, a major part of the data collected in the present

study deals with teachers' diagnostic agreement.

Case simulation

 

Observational, computer simulation, and training
studies of the clinical encounter and the tenets of the
Inquiry Theory all involve the use of simulated cases of
a client's problems. (A case is a set of information that
the clinician can collect). These simulated cases are
based on actual clients and represent the relevant
characteristics of the client's problems.

Two types of simulated cases have been develOped:
manual based and computer based. The manual based case
consists of various kinds of information about a child's
reading performance (e.g., background data, school records,
results of different tests that the child took, etc.) which
are stored in a box and retrieved manually. The computer
based case involves the same information, but it is stored
in a computer file and is retrieved via computer terminal.

The use of simulated cases in studies of reading
diagnosis is desirable for several reasons: (1) they
provide a means for presenting the child's behaviors to
the clinicians; (2) they allow for replications of the

clinical encounter with the same case (and with the same or

63
different clinicians), so that an acceptable level of
objectivity can be achieved; (3) they may prove to be a low
cost tool for training clinicians in diagnosis and remediation
skills; and (4) they may prove to be an effective tool for
training clinicians and educators on diagnostic skills
under immediate feedback.

Following is an examination of the three types of
studies designed to investigate the clinical problem solving
behavior of reading specialists and classroom teachers in
reading diagnosis. All three are guided by the principles of
the Inquiry Theory and utilize simulated cases as an
experimental tool. The studies which will be reviewed are
observational studies, computer simulation and training

studies.

Observational studies

 

Several observational studies, other than the
present one, were conducted during 1977 and 1978. The
prime objective of the studies was to investigate the
clinical problem solving behavior of reading specialists
and learning disabilities personnel as they diagnosed
children's learning difficulties. The studies used
manually based simulated cases of learning difficulties.
The general procedure was as follows (Gil, Vinsonhaler
and Wagner, 1978):

1. An Observational session was held in which the clinician
(reading specialist or learning disabilities personnel)

interacted with a simulated case of learning difficulties

64
to reach diagnostic decisions about the case. The
clinician was given an inventory of information about
the case, and was instructed to ask for any piece of
information that he/she wanted. An experimenter
retrieved the items requested and handed them to the
clinician. As the clinician examined this information,
the experimenter encouraged him/her to talk aloud and
verbalize his/her thinking. In the meantime, a
clinical observer (a reading specialist or a teacher
with a reading background) sat behind a one-way mirror,
managing recording equipment and keeping a precise
record of all information requested, the time of each
request and the clinician's verbatim comments.
Immediately following the Observational session, the
clinician was instructed to write down his/her diagnostic
judgments about the case and to map out a general plan
for remediation.
A debriefing session was held, during which the
clinician was asked to recall his/her problem-solving
behavior in requesting and evaluating information about
the case.

Each clinician interacted with two simulated cases;

the second case was either a replication of the first or

a different case. Data analysis consisted of both process

and product measures. Process measures included such

variables as number of hypotheses generated by the clinician

during the session, number of cues (items of information)

requested, sequencing, and orgnization of the information.

65

Product measures focused on diagnostic consistency (how
consistent the clinician was in describing reading per-
formances across two cases) and diagnostic commonality
(how closely the clinicians agree with themselves and with
each other in their diagnoses). Also included in the
product measures was a comparison of each clinician's
final diagnosis and the criterion diagnosis written by
an expert clinician.

Preliminary results from the first study (Vinsonhaler,
1979) indicate that agreement is quite low, both between an
individual clinician's performances on alternte forms of
the same case, and between his/her diagnosis and those of
other clinicians. This means, then, that the reliability of
a clinician's diagnosis is low. Furthermore, diagnostic
consistency does not appear to be a function of the time
spent on a case. Rather, consistency seems to vary with the
number of replicate cues (items of information) that a
clinician examines across cases. This suggests that
clinicians may use a heuristic or a systematic data

collection plan across cases.

Computer simulation studies

 

Computer simulation studies are another means by
which knowledge of effective clinical problem solving can
be gained. According to the Inquiry Theory, the behavior
of a clinician is a function of his/her memory and strategy

(Vinsonhaler, Wagner, and Elstein, 1977). Therefore, a

66
simulated clinician can be created by programming into a
computer a memory, a strategy, and the interaction between
the two, where changes in memory, in strategy, or in both
will be reflected in changes in the simulated clinician's
performance.

In computer simulation studies, a simulated clinician
interacts with a simulated case of a child with reading
difficulties (Gil, Wagner and Vinsonhaler, 1979). Through-
out this interaction, the computer uses its memory and
strategy to arrive at a diagnosis for a child. In so doing,
it uses a process similar to the one applied by human
clinicians. The interaction between the computer clinician
and a case starts when the computer receives the "initial
contact" about the case, which is some basic information about
the case's reading status. With this information, the
computer proceeds to collect more information, on the basis
of which it generates hypotheses about the child's reading
problems. Then the simulated clinician collects still
more information to confirm or disconfirm these hypotheses.
After collecting a certain amount of information about the
case, the simulated clinician then begins its final diagnosis.

Different memories and strategies may be used to
create different simulated clinicians, which may interact
with various cases to reach diagnostic decisions. Analysis
of the effects of these diagnostic outcomes would increase
our understanding of the clinical diagnostic process, and
the information could then be applied to the training of

human clinicians.

67

Training studies

 

The third kind of study based on the principles of
the Inquiry Theory is the training or application study. The
prime objective of such studies is to examine the application
of clinical problem solving theory in the training of
reading specialists and classroom teachers. (The
instructional corollary from which this objective is
derived states that if clinical performance is determined
by the clinician's memory and strategy, then this performance
may be improved by alterations in clinical memory and
strategy).

One training study (Sherman, 1978), carried on
during the summer of 1977, addressed the following
questions: (1) can teachers be trained to diagnose reading
difficulties in a manner similar to that of effective reading
specialists? (2) does the instructional corollary hold for
reading diagnosis?

To answer these questions, the performance of 36
students enrolled in a graduate diagnosis course was
examined. Instruction was explicitly guided by the
principles of the Inquiry Theory. Pre and post tests
consisting of two instruments were administered: First,

a memory battery association test examined the students'
memory of problems in reading. It measured how well the
students matched information to cue association (e.g.,
located specific findings about a child's reading per-

formance); how well they matched cue to factor association

68

(e.g., identified important strengths and weaknesses given a
set of findings), and how well they matched factor to cue
association (identified important cues in testing hypothe-
sized strengths or weaknesses).

The second instrument was a diagnostic performance
test, which examined the students' ability to accurately
diagnose simulated cases of reading difficulty. This test
consisted of a set of materials designed for use with
simulated cases; included in the test were standardized
instructions, response forms, and a scoring key, each
intended to facilitate the student's interaction with the
simulated cases.

In addition to the traditional presentation of
course content by the instructor, the students interacted
'with manual based and computer based simulated cases of
reading difficulty, using the materials from the diagnostic
performance test. At the end of the course, the students
took the memory battery association posttest, and the
diagnostic performance test (as they interacted with
different simulated cases).

Results indicated that the students did learn to
diagnose in a manner similar to that of reading specialists;
there was a significant increase in their mean diagnostic
score. Results also suggested that the learning of clinical
memory seems to be related to the learning of clinical

diagnostic performance.

69

Clinical studies of reading diagnosis: Conclusions

 

The Inquiry Theory presented in this section and the

three kinds of studies derived from it suggest the following:

1.

Clinical Problem solving behavior seems to be determined
by the clinician's memory, strategy, and the interaction
between the two. Therefore, clinical problem solving per-
formance can be improved by providing clinicians with
external and training aids to memory and to strategy.
Clinical problem solving behavior can be simulated by
computer; hence, by altering memory and strategy, simulated
clinicians can be created which represent both typical and
ideal approaches to problem solving (e.g., with or without
heuristics). Human clinicians can then be trained to
utilize those problem solving approaches which are found
to be effective in the computer simulations.

Case simulation appears to be a useful, effective, and

low cost technique for training reading specialists and
classroom teachers in diagnostic skills.

Educators can be trained to make more precise diagnostic
judgments given proper training involving simulated cases

with feedback.

Summary

An attempt has been made in this chapter to review the

literature dealing with two subjects: 1) the nature of

reading diagnosis and its implications for classroom teaching,

and 2) recent studies of clinical problem solving behavior

70

in reading diagnosis upon which the present study is drawn.

Examination of the literature in the first area
focused on the importance of reading diagnosis, various ways
of defining reading diagnosis, different levels of reading
diagnosis, and diagnosis made by classroom teachers,
including analysis of some pertinent research studies. It
concluded with description of Barr's Model of Diagnosis and
Sherman's Model of Reading and Learning to Read, which
represent two recent attempts to deal, in a more systematic
way, with the content and process of reading diagnosis.

The conclusions drawn from this examination of the
literature about the nature of reading diagnosis were that:
1. Reading diagnosis is regarded by most educators as an

integral part of all teaching;

2. Early and continuous reading diagnosis is essential to
prevent reading problems from becoming more severe;

3. Various definitions are attached to the term "reading
diagnosis", and the description of levels of reading
diagnosis differs from one source to another;

4. The literature fails to describe or to analyze the cog-
nitive processes performed by teachers as they diagnose
children;

5. Although many educators call upon classroom teachers to
assume the responsibilities of diagnosing and remediating
children's reading difficulties, there is a lack of
empirical research which might determine teachers'

effectiveness in diagnosis and remediation skills.

71

The review of recent studies of clinical problem

solving behavior in reading diagnosis began with a general

description of the "Inquiry Theory" and of the utilization of

simulated cases in studies of clinical problem solving

behavior. It continued with brief description of three kinds

of studies derived from the tenets of the Inquiry Theory --

Observational, Computer simulation and Training -- all of

which use simulated cases as a research tool.

The conclusions drawn from the review of these

studies were that:

1.

the

Clinical problem solving behavior is a function of the
clinician's memory and strategy.
Clinical problem solving behavior can be simulated by
computer;
Simulated cases can be used effectively in studies of
clinical problem solving behavior, and
Educators can be trained to make more accurate diagnostic
judgments through interaction with simulated cases.

The implications of this review of the literature for
present study are as follows:
If classroom teachers are regarded as diagnosticians, it
is essential that extensive research be conducted on
teachers' diagnostic skills.
Such research should focus on the mental tasks performed
by the teachers as they diagnose cases, so that we will
begin to understand the diagnostic process; and
Simulated cases of reading can be used as research tools

in studies of teachers' diagnostic skills and processes.

CHAPTER III

DESIGN AND METHODOLOGY

Introduction

 

The purpose of this study was to investigate the
Clinical Problem Solving Behavior of classroom teachers.
There were two main Objectives:

1. To examine the clinical problem solving behavior of
classroom teachers as they diagnosed children's reading
problems.

2. To explore the similarities and differences between
teachers' diagnostic practices in experimental and

classroom situations.
The Subjects

The subjects of the study were 10 elementary school
teachers (all females) who volunteered, for payment, to
participate. Because data collection was a lengthy procedure,
and because it involved both experimental and classroom
situations, the number of the subjects was limited to 10.

For this reason, and because of the exploratory nature of the
study, a deliberate effort was made to select teachers with
different backgrounds to ensure diversity of performance.
Five teachers were chosen from a group of teachers trained

72

73
in'a reading diagnosis course taught by Dr. Rebecca Barr
at the University of Chicago; the other five were from the
Lansing area.

Chicago teachers have all taken a reading diagnosis
course taught by Dr. Rebecca Barr. This course trains
teachers to diagnose students' skill level in reading by
observing the children's reading behavior and by listening
and questioning them. Teachers are given materials which
would be available to them in a regular classroom setting and
are asked to interact with cases of reading difficulties.

The five teachers who participated in this study were selected
from a group of eight teachers. Two criteria were used for
selecting them: 1) number of years of classroom teaching,

and 2) guaranteed employment for the following fall. The
second criterion was necessary because the second phase of

the study involved interviews with the teachers in their own
classrooms.

' Lansing teachers: Three of the five Lansing teachers
were also participants in another study in progress at the IRT
-- the Teachers Conceptions of Reading study, coordinated by
Dr. Gerald Duffy. These teachers were asked to participate in
the present study during a meeting of all the teachers taking
part in Dr. Duffy's project. A fourth teacher was recommended,
and invited to participate, by a member of the Clinical
Studies group. The fifth teacher was recommended by an

experienced teacher known to the experimenter.

74
To participate in the study, all teachers had to have
completed at least two graduate courses in reading instruction,
preferably one in reading diagnosis.
The teachers' backgrounds, schools and grade levels

taught are summarized in Table 2.

Design

The design of the study called for two phases of
research: Phase 1 was conducted during the summer and early
fall of 1978 under laboratory conditions. The teachers were
asked to interact with simulated cases of reading difficulties
and to arrive at diagnostic judgments about the cases'
problems.

In Phase 2, carried out during the fall of 1978,
the teachers were observed for one reading session in their
own classrooms, and then interviewed about reading instruction
and diagnostic practices.

The overall design of the study is illustrated in

Table»3.

75

 

 

 

 

 

 

 

 

mm ~\H mmmao manta: .m.z ca me n
mm H mmmao manta: .4.2 ha mm o
ma m mmmHO manna: .B.m.z H mm m
.a.m.z
ow m mmmao wanna: .<.z m Hm v
manoeuiounosm
\Hmsmcflaflm
mm m Immoao 30A .B.w.2 ha we m
mauwm
Ham\oflaonumu
NH m Immmao wanna: .B.m.z m mm N
Hoonom non
mm N Immwao OHOOHZ .m.z ma me a
mmmao cw mcanommu waucon mocmwnmmxm
mucmpsum luau nown3 ca mcasomou
mo .02 Demon» ocmuo Hoonom wo mama Odo: omumoo mumom mo .02 O04 muocomoe

 

 

 

 

 

pcsoumxomn mopmum Ocm Hoosom .muonommell.m mummy

 

 

76

m.mN

Humemum sum
mlmnmucoEme
momma
nlhumucmfiwam
Hm3oq

Huamsmanann\zoa
HIMMMHU
maeeasusoa
Humauam

Ham oaaoaumo
\mmmHo mavens
enema

\mmmao maven:
mummmHo mavens

mumafism
no cows

 

HN

mmmao mavens

mm

ca

 

NN

mmmao OHUOHS

mm

 

NN

mmmao maennz\3oq

AH

Nm

 

mmmHO cw
mucmcsum
MO .02

 

pcmsmv OUOHO

 

mcflsowou maucou
IHDO cOH£3 cw
Hoosom mo mama

 

pawn moumoa

ewsaflucoouum mamas

 

OOOOHHOQRO
mcwnommu
mums» mo .02

 

mmfl

 

mumcomme

77

 

 

 

 

 

 

 

 

 

 

 

mchmou
OH
mmOHuomnm 3OH>HODOH 0H
HOHOOEOH conmom .um H ume m
can “mom muouco>cH w m
OHumOOOMHO DHEHH m HO < Show once a w m
.HchHu GOHmmmm mmmo mEHu huoucm>cH can once m m
nonhumcH mcwpmmu nonuocm oz .um m ume pmumHaeHm m m
.wnwnowou mco mo cqu H m no u so v o
on» usonm chHum> OOHmmmm monmom maloan3 5 .Oz mmmu mchHmnB m 4
m3OH>noucH luomno mo mch OOHDMHmeom ”conmwm Hm N m
EooummMHu mchuoz .MOHHmmm uOHHan a mHmocmea IGOHum>Homno .HumcH Hmuocmw H a
monB¢> N H
m3MH>mmezH ummmmo onmmmm onmmmm
mhmH HH<m mhmH MHDO I mzoh "onBNDBHm

ZOHB<DBHm EOOMmmdqu MfiOBflmomdn ZH mmmdo QMBflHDZHm m0 MHwozwfiHQ

 

mcsum may mo cmHmmc HHmum>OI|.m mHmme

78

Phase 1: Laboratory conditions

 

Each of the teachers visited a reading clinic at
two occasions. The Chicago teachers went to a clinic at the
University of Chicago, the Lansing teachers reported to a
clinic at the College of Education at Michigan State University.

At each session, the teacher was asked to interact with
a simulated case and to arrive at diagnostic decisions about
its reading difficulties.

The two cases used were developed by Dr. Rebecca Barr
and were based on two Chicago elementary school students.
They were chosen by Dr. Barr for this study because both
had problemsthat were representative of reading difficulties
encountered in the public schools by classroom teachers. One
of the students was an eight-year-old girl; the other a
10 year-old girl.

Both cases were built around materials available to
classroom teachers for reading instruction and the
children's natural language patterns. As such, each case
included the following: 1) a copy of the passage that the
child read orally; 2) a tape-recording of the child reading
the passage; 3) an oral transcription of what the child
read; 4) background information about the child and the
school, and 5) transcripts of the child's performance on
various tasks related to the reading passage, such as her
answers to comprehension questions; her definition of a word

presented to her, and her ability to identify a word printed

79

on the board (See Appendix A for the Case Information

Inventory).

The sequence of steps followed during this phase of

the study was as follows:

1.

A training session was held, during which the teacher
was given an introduction about the purpose of the study.
Instructions about the procedures to be followed were
read, and each teacher was asked to practice with a
training case and a training case inventory. (This
step was essential to ensure that all teachers
understood what they were expected to do, what
information in the Case Inventory was available to
them, and how they were to use the Case Inventory to
retrieve and receive the desired information.) (See
Appendix B, Directions for Observational Study.)

A one-hour observational session was held, during which
the teacher was given an inventory of information
available for the case and was instructed to ask for
any information she wished to have. The use of a

Case Inventory was necessary for two reasons. First,
it let the teachers know what specific units of
information were available. Second, it helped
standardize procedures, including the information
available, which in turn helped standardize the
teachers' memories. This allowed for a more objective
analysis of the data than would have otherwise been

possible.

80
As each teacher examined the information about a
case, she was asked to verbalize her thinking. The
experimenter who has instructed the teacher retrieved the
information requested and handed it to the teacher.

Teachers were allowed to keep all items of information

requested throughout the session. The experimenter also

encouraged the teacher to talk aloud as she examined the

information. In the meantime, a clinical observer (a

reading specialist) sat behind a one-way mirror, managing the

recording equipment and keeping a precise record of all
information requested, including time of each request and

the teacher's verbatim statements.

3. Immediately after the observational session, the
teacher was instructed to write down (1) her diagnostic
judgments about the child's reading problems and (2)

a general plan for remediation. A time limit of half
an hour was set for this activity.

4. A debriefing session (with no time constraint) followed,
during which techniques of stimulus recall were used
to help the teacher remember her problem—solving
behavior in requesting and evaluating information about
the case. The clinical observer helped the
experimenter conduct the debriefing session.

The teacher was shown each of the items of
information she had requested during the Observational
session. A standard set of questions about the items was

asked to explore and clarify her problem-solving behavior.

81

A second interaction between the teachers and a
different case took place several days later. The same
procedure was followed as for the first case. In addition,
after the debriefing session for the second encounter, a brief
demographic questionnaire dealing with the teachers' training
experience and teaching experience was administered.

All 10 teachers interacted with the same simulated
cases, but in different orders. Five interacted with Case No. 7
first and Case No. 8 second; the others interacted with Case
NO. 8 first and then with Case No. 7. Two inventory forms
were available for each case. The forms were identical, except
for the order in which they listed the information about the
student. The teachers were given Form A for one case, and
Form B for the other. The distribution of cases and inventory

forms is summarized in Table 4.

Phase 2: Classroom interviews

 

During the fall of 1978, each teacher was visited once
in her own classroom. There she was observed for one reading
session and interviewed about her reading instruction and
diagnostic practices (See Appendix C for Classroom Interview
Form).

The question on which the interview focused was: How
do teachers diagnose children with reading difficulties in the
natural classroom situation? Specifically, questions were

asked which were designed to explore the teachers' memory

82

TABLE 4.--Assignment of cases and inventory forms*

 

 

 

 

 

 

 

 

 

 

 

 

SESSION 1 SESSION 2
TEACHER
Case Inventory Case Inventory

1 8 A 7 B
2 7 B 8 A
3 8 B 7 A
4 7 A 8 B
5 8 A 7 B
6 7 B 8 A
7 8 B 7 A
8 7 A 8 B
9 8 A 7 B
10 7 B 8 A

 

 

 

 

 

 

*For analysis purposes, the cases were identified as Number
7 and 8, respectively, rather than by the children's names.

83
and strategy with regard to reading diagnosis.

The interviews proceeded in the following manner:
The experimenter handed the teacher four descriptions of
real children with typical reading problems. Two of the
descriptions were based on the simulated cases the teacher
had diagnosed in the laboratory. The teacher was instructed
to read these descriptions and think of children in her own
classroom who corresponded to them. After she identified
such children, she was asked to indicate which parts of the
descriptions did or did not fit the selected children. This
explanation provided the experimenter with a more precise
idea of the resemblance between the children in the classroom
and those in the description.

The teacher was then asked a set of questions, which
explored how she identified children with reading problems
and how she diagnosed the problems. Each teacher was asked
to describe diagnostic procedures for two children. (See
Appendix D: Descriptions of the four children. Also see
Appendix E, which shows the frequency with which teachers
indicated similarities between descriptions and children in

their classrooms.)

Data Analysis

 

Data analysis included two major types of measures:
Product and process (Wagner et al., 1979). Product measures

dealt with the outcomes of the clinical encounter between

84

teacher and case, and included the final stated diagnostic

judgements, what cues were collected, etc. Process measures

focused on the manner in which the problem was diagnosed,

e.g., on relationships between cues collected and statements

made about the cues. Analysis was also conducted based on

the classroom interviews, to compare teachers' diagnostic
processes in both laboratory and classroom situations.
Product measures involve three types of statistics

(Vinsonhaler, 1979):

1. The proportional agreement, which is a measure of group
agreement. This measure is based on the final stated
diagnoses written by all teachers for a given case, or on
the list of cues collected by all teachers for a case,
and is determined in the following manner: First, a do-
main of statements (stated diagnoses or cues collected)
is defined. Next, the percent of teachers who arrived at
the same diagnoses or collected the same cues about a
case is calculated. The purpose of this measure is to
indicate which diagnoses or cues are most frequently
mentioned or collected by the teachers.

2. The commanality score, a measure of how closely an
individual teacher and a defined group of teachers agree
in terms of diagnostic statements or cues collected for
the same case.

3. The inter-clinician agreement correlation, which is a

measure of the agreemnt among teachers on the diagnostic

85
statements or cues collected for a given case. This
statistic is calculated as follows: (1) a domain of
statements (written diagnoses or cues collected) is
defined for a case; (2) those statements appearing in an
individual teacher's interaction with the case are
identified; and (3) a two by two contingency table is
prepared and a Phi coefficient is calculated; the cells
in the contingency table include (a) the number of
statements mentioned in both encounters; (b) statements
present in the first encounter (one teacher) but not in
the second; (c) statements present in the second
encounter (second teacher) but not in the first; and (d)
statements not present in either encounter. (See
Appendix I: Observational Studies Data Analysis System
[OSDAS] statistics).

Process measures of the clinical encounter between a

teacher and a case were made which indicated the manner, or

diagnostic process, by which the case was diagnosed. Four

measures were selected for analysis in this study:

1.

Length of interaction: amount of time (in minutes) that
the teacher interacted with the case during the
observational session. Time was measured from the point
at which the teacher received the initial contact in-
formation about the case until she indicated she was
ready to write the diagnosis.

Number of stated diagnoses: When the teacher indicated

86

she was ready to write the diagnosis, she was allowed
half an hour to write down her diagnostic statements
about the case.
Number of cues collected during the observational session.
The teacher was allowed to look at the case inventory and
collect as many cues as she wished. However, she was
required to ask for them one at a time. This condition
will allow the experimenter to conduct an exploratory
investigation of the order in which the teacher collected
information about the case.
Relationships between cues and statements: During the
observational session, it will be recalled, the teacher
was asked to verbalize her thinking as she collected in-
formation about the case. This "thinking aloud" continued
during the debriefing session. During both sessions a
clinical observer kept a precise record of the teacher's
verbatim comments about each of the cues she collected.
It was thought that the relationships between the
teacher's verbatim comments and the cues she collected
might shed light on her diagnostic processes.
Accordingly, a lO-point scale was devised for
characterizing the relationships between the teacher's
statements (verbatim comments) and the cues:
l) The cue requested confirmed a statement, for example:

"It confirmed what I suspected that a general weakness

exists."

2)

3)

4)

5)

6)

7)

8)

9)

10)

87
The cue requested disconfirmed a statement, for
example: "I ruled out that concepts are a major
problem."
The cue requested neither confirmed nor disconfirmed
a statement.
The statement deals specifically with the cue, for
example: "She doesn't use whole words."
The statement is a generalization of findings about
cues, for example: "Comprehension skills are more of
a problem for her (than I thought)."
The statement is related to cues in a manner other
than the above, for example: "I don't know why she
didn't understand."
The statement indicates the teacher's method of
collecting information about a case, for example: "I
wanted to probe the issue of word identification."
The statement made has no relation at all to the cue,
for example: "Would be a help to me if it wasn't
quite so challenging to her."
The statement is about remediation, for example:
"Should go back to easier level."
The statement is about children's reading behavior in
general, or about the teacher's philosophy of reading,
for example: "sometimes re-reading helps kids to get

more information."

88

Comparisons between teachers' diagnostic processes in

 

laboratoty and classroom situations.

 

Two means of validating the teachers' performance on
a simulated case were used in this study. First, the teachers
were asked to interact with a second case, while the
investigator looked for consistency in performance across the
cases; and second, the investigator studied the diagnostic
processes teachers employed in their natural classroom
situation and looked for similarities and differences between
these processes and the diagnostic processes exhibited in the
laboratory setting..

This second step was very important, because it may
indicate the degree to which teachers' interactions with
simulated cases in the laboratory setting reflect their
natural diagnostic processes. It should be emphasized, how-
ever, that the laboratory phase is essential if one wants to
trace the diagnostic processes employed by teachers.
Observations of systematic cue collection and "thinking aloud"
are not available in a classroom situation.

Comparisons of the diagnostic processes used in
laboratory situations and those used in the natural classroom
setting centered around the following questions: What are
the diagnostic categories most frequently mentioned by
teachers in the classroom? Are there similarities between

the diagnostic categories mentioned by teachers in the

89
laboratory situation and those mentioned in classroom setting?
Questions relating specifically to classroom inter-
views were also examined, among these were: What are the
remedial approaches offered by teachers in the classroom?
How successful do teachers perceive themselves to be in

remediating children's reading difficulties in the classroom?

Summary

This study was conducted to investigate the clinical
problem solving behavior of classroom teachers in diagnosing
reading problems in laboratory and classroom situations.
Research was divided into two parts: (1) observation of
teachers' interactions with two simulated cases in a
laboratory setting, and (2) interviews with the same teachers
in their classrooms.

Two major types of measures were used to analyze the
data: product measures and process measures. Product
measures were concerned with the outcomes of the clinical
encounter between a teacher and a case. Process measures
were concerned with the manner in which the problem was
diagnosed. Comparisons were also made between teachers'
diagnostic practices in the laboratory situation and their
practices in the natural classroom environment.

The data which were collected, treatment of the data,

and results are presented in the next chapter.

CHAPTER IV

RESULTS

Introduction

 

This study was designed to investigate the clinical
problem-solving behavior of classroom teachers as they
diagnose reading performance in laboratory and classroom
situations. The study involved 10 teachers from the
Chicago and Lansing areas and was conducted in two phases.
In the first part, the teachers interacted with two simulated
cases of reading difficulty in a laboratory situation.
Techniques of stimulus recall were used to obtain recorded
verbatim of the teachers' diagnostic processes (written
protocols of the teachers' "thinking aloud" were also taken,
as were their final written diagnoses about the case). In
the second phase of the study, the teachers were observed
in their own classrooms and interviewed about their
instructional, diagnostic, and remedial practices. The
interviews were audiotaped and transcribed.

Because few investigations have been done of the
clinical problem solving behavior engaged in by classroom

teachers as they diagnose reading performance, no hypotheses

90

91

were established for this study. Rather, the study was
viewed as an exploratory one, intended to yield a
description of the teachers' diagnostic processes.

Data analysis focused on the degree of similarity
(and difference) among the memories and models the teachers
used in diagnosing the cases of reading difficulties. The
theoretical framework on which data analysis was based was
the Agreement Corollary, derived from the Inquiry Theory of
Clinical Problem Solving Behavior (Vinsonhaler, Wagner and
Elstein, 1977).

The Agreement Corollary, it will be recalled, states
that, if diagnosis is determined by clinical memory and
clinical strategy, then the use of similar memories and
strategies will result in similar diagnoses. Memory and
strategy are determined, in part, by the training clinicians
receive; the more similar the training they receive, the
more similar their memories and strategies will be, and the
more closely they will agree with each other when diagnosing
the same case of reading difficulties. Conversely, if
training varies from one teacher to another, the teachers
will operate from different "memories" or "models" of the
reading process and will tend to show limited agreement in
their diagnoses (for a more complete discussion of the
Inquiry Theory and the Agreement Corollary, see Chapter II,

pp. 57—62).

92

For purposes of clarification, recall the two
teachers, discussed in Chapter II, who employ different
memories or models. One, it was said, Operates from a
skill-model orientation, the other from a socio-
psychological model. Because of their different diagnostic
backgrounds, they tend to entertain different diagnoses
about a case of reading difficulties. The following is a
list depicting some of the diagnoses the two might
entertain for the same case (in other words, this is a
list of some of the diagnostic categories "stored" in each
teacher's memory, based on, and derived from, the training

teachers received):

 

 

Skill orientation Socio-psychological orientation

poor phonics* low interest in reading*

poor word analysis poor visual acuity

poor oral reading* reads below grade level*

poor comprehension poor auditory acuity*

poor sight words* inadequate verbal ability

reads below grade level* no intellectual models of reading
at home

The * indicates the final stated diagnostic judg-
ments reached by each teacher about the case. As seen,
they differ significantly. A careful look at the
judgments indicates that each teacher mentioned only some
of the child's problems. The skill-oriented teacher, for
example, cites "poor phonics" as a problem area, but does

not identify a possible cause for this problem, which may

93

be "poor auditory acuity" (mentioned by the second teacher).
The first teacher also fails to identify the cause for the
child's "poor sight words," which may very likely be "low
interest in reading" (mentioned by the second teacher).

Statistcal procedures -- such as proportional
agreement statistic, commonality score and inter-clinician
correlations -- can be used to calculate the teachers'
diagnostic agreement [See Appendix I: Observational Studies
Data Analysis System (OSDAS) Statistics]. An inter-
clinician Phi coefficient calculated to find the diagnostic
agreement of the two teachers in the example, yields a
score of around 0.0. This means that the two teachers
show almost no agreement with one another about the child's
problems. Out of a total of seven diagnoses mentioned by
the two teachers, only one ("reads below grade level") is
made by both. This limited agreement can be attributed to
the teachers' different training background, which have
left them with different "memories" about reading and
reading deficiencies.

In diagnosing the child's performance, each teacher
considered only those factors that could be derived from
the narrow list of categories stored in his/her memory.
Thus, neither teacher's diagnosis, by itself, is reliable
enough to determine the remedial plans to be taken for the
child. A combination of their diagnoses would probably
yield a more accurate, comprehensive, and reliable diagnosis,

which would lead to a more effective remediation.

94

As mentioned, the Agreement Corollary also applies
to the strategies teachers use. In the realm of reading
diagnosis, there may be two general and distinct strategies,
or ways of reasoning: deductive and inductive. For purposes
of illustration, recall the two teachers in the example, one
of whom operates from a skill-model orientation and one of
whom works from a socio-psychological model. Both used the
same initial informal reading assessment procedures to
evaluate the student's reading performance (e.g., examination
of school records, analysis of audiotapes of the child
reading a passage, etc.). They differed, however, in their
subsequent cue collection. The skill-oriented teacher,
being an inductive reasoner, tested the child on various
skills associated with reading, such as the child's ability
to chunk, analyze, and identify words. After collecting
the information about the case, this teacher concluded that
the child lacked sufficient reading skills. The teacher
Operating from the socio-psychological orientation, on the
other hand, was a deductive reasoner; she/he assumed that
the child's problems were emotionally-based and thus
collected only that information pertaining to the child's
personality, such as psychological reports, social-worker
comments about the child, and home background.

In summary, it may be stated that due to inadequate
and/or insufficient training, different teachers may

Operate from very distinct, and narrow, models of reading

95

and the factors that affect reading performance. When

diagnosing a given case of reading difficulties, they may
operate from different memories and strategies, which will
result in very limited agreement about the case's reading

performance.

Presentation of Results

 

The purpose of this chapter is to present the
findings of this study, most of which relate to the issue
of agreement among teachers. These findings will be
presented in three parts;

1. Product measures - which were concerned with the
outcomes of the clinical encounter between a teacher
and a case,

2. Process measures - which were concerned with the manner
in which the problem was diagnosed, and

3. Classroom interviews - which were concerned with teachers'
instructional, diagnostic and remedial practices in the

natural classroom environment.

Part 1: Product measures

 

Product measures, which were concerned with the
outcomes of the clinical encounter between a teacher and
a case, dealt with the teachers' final stated diagnoses
and the cues they collected for a given case. The

measures were divided into three parts:

96

l. Proportional agreement - a measure of group agreement
on diagnoses and cues collected;

2. Commonality score - a measure of the agreement between
individual teacher and a defined group of teachers in
terms of diagnostic judgments and cues collected on
the same case, and

3. Inter-clinician agreement correlations - a measure
of the agreement of one teacher with another on the

diagnostic judgments or cues collected for a given case.

The proportional agreement statistic.--the pro-

 

portional agreement statistic was calculated by dividing

the number of teachers who stated the same final diagnosis
(or collected the same cues) for a given case by the total
number of teachers who interacted with that case. This
process yielded the percentage of teachers who arrived at

the same diagnostic judgments or collected the same cues
about the case. It also indicated the most frequently stated

diagnoses or cues collected.

Prgportional agreement for diagnostic judgments.--

 

before any statistical operations were carried out, a list
of total diagnostic judgments was prepared. Included were
all diagnostic judgments mentioned by all 10 teachers for
each case. Categorizing the judgments turned out to be
somewhat of a problem. Many of the judgments were stated

differently, although they were essentially the same,

97

because there is no generally agreed upon vocabulary for
reading diagnosis. To determine which of the differently—
stated judgments were, in fact, identical, a senior reading
clinician looked at the diagnostic list and omitted all
redundant categories. For purposes of reliability, another
reading clinician did the same thing, and his list of
diagnostic categories was checked against the first clinician.
The result was a list of 54 diagnostic categories for Case 7,
and 62 for Case 8 (See Appendix F for the diagnostic domain
for Case 7 and 8). Proportional agreement statistics were
then calculated for the most frequently-mentioned judgments.
Table 5 shows these statistics.

The first column in the table lists the diagnostic
judgments mentioned by at least 20 percent of the teachers
for a given case. The second and third columns indicate the
percentage of teachers mentioning each category for Case 7
and Case 8, respectively. For example, the diagnostic
category "Endings: Ignores", was mentioned by 50 percent of
the teachers for Case 7, but not at all for Case 8.

As the table suggests, in general, the teachers showed
a very low level of agreement in diagnosing the cases. Only
one diagnostic category for each case was mentioned by five
teachers (50 percent of total), and only one, for Case 7,
was mentioned by four teachers. Seven categories were
mentioned by three teachers, and 16 by two. The rest of the

judgments were mentioned by only one teacher.

98

TABLE 5.--Most frequently mentioned diagnostic categories

for Case 7 and Case 8 and 10 teachers

 

SELECTED DIAGNOSTIC CATEGORIES

Comprehension: Poor

Comprehension: Adequate

Understanding: Good when listening

Listening comprehension: Adequate

Oral reading comprehension: Good

Vocabulary concepts: Adequate

Vocabulary: Weak

Oral reading: Skips words

Oral Reading: Does not pay attention
to punctuation

Puncuation: Lacks

Endings: Ignore

Sight words: Weak

Sight words: Good

Reversals: Problem area

Mispronunciation errors: Did not correct

Phonic skills: Weak

Beginning sounds; Knows

Initial consonants: Strong

Syllabication skills: Poor

Unfamiliar words: No attempt to sound

Vowels: Need work

Word attack skills: Lacks

Context to help with the unknown word:

Does not use

Word analysis: Weak

Language: No language problems

Print translation: Problem area

CASE 7
(54 total
DX)
.30
.20

.30

CASE 8

(62

total
DX)

20
20

20
30

.30

.20
.30

 

*—-denotes 0.00

It should also be noted that only one diagnostic

category (”Vocabulary concepts: Adequate") was mentioned by

as many as three teachers for both cases.

All other

categories were mentioned either for Case 7 or Case 8, but

not for both.

99

Figure 1 shows the percentage of diagnostic categories
in the entire diagnostic domain that the different teachers
mentioned. The verticle line indicates percentage of total
diagnostic judgments made; the horizontal line indicates the
number of teachers making these percentages of total judgments.

The figure reveals limited agreement among the
teachers on the percentage of total diagnostic judgments they
made for both Case 7 and Case 8. For example, 80 percent
of the total diagnostic judgments made for Case 7 were made
by only one teacher, whereas only 2 percent of the total
were made by five teachers. For Case 8, 73 percent of the
total judgments were made by only one teacher, and only 1.5
percent were made by five teachers.

In summary, the data presented in Table 5 and in
Figure 1 show little agreement among the teachers on

diagnostic judgments for either of the two cases.

Proportion agreement for cues.--the proportion

 

agreement for cues was calculated in a manner similar to
that used for determining the proportional agreement for
judgments. First, a list of all cues collected by all
ten teachers for each case was prepared. All redundant
cues were eliminated, and a frequency count was made to
determine how many teachers collected each cue.

Table 6 shows the proportional agreement statistics

for the most commonly collected cues for Cases 7 and 8.

100

100.
95‘ _ . , ,
J . . case 7 (N=34 diagnostic judg-
90 ments) mean = 1.45
85' x ----- x case 8 (N=62 diagnostic judg—
80. ments) mean = 1.34

75.

70-
65-
60-
55-
50‘

40‘
35-
30'
25-
20 ‘
15 ‘
ioI

Percent of diagnostic judgments mentioned

 

 

 

y t I i 1

1 2 3 4 5 6 7 8 9 10

Number of teachers mentioning category

Figure l.--Percent of diagnostic categories for Cases 7 and
8 and 10 teachers

101

TABLE 6.--Most frequently collected cues for Cases 7 and 8

and 10 teachers

 

SELECTED CUES CASE 7 CASE 8
(221 cues (321 cues
total) total)
Background: Cultural/Home .90 .90
Background: Student's personality -- .90
Background: Comments on student reading
the passage .70 .60
Background: Test information .50 .70
Vocabulary concepts: "Fancy" (in isolation).70 --
Background: Reading program .60 .40
Background: Language development -- .60
Word recognition: "Simple" (in isolation) .60 --
Vocabulary concepts: "Detective"
(in isolation) -- .60
Background: Oral reading transcription .50 .40
Background: Performance in school .50 --
Background: Picture accompanying the
reading passage .50 --
Vocabulary concepts: "Fancy" (in teacher
sentence) .50 --
Word recognition: "Sammy" (teacher chunk) .50 --
Comprehension: Oral reading (paragraph 1,
list) -- .50
Comprehension: Oral reading (paragraph 1,
question 10) -- .50
Word recognition: "Everyone" (in isolation) -- .50
Word recognition: "Detective" (student
chunk) -- .50
Background: School history/description .40 .40
Background: Description of picture in
the passage .40 --
Comprehension: Oral reading (no paragraph
specific, list) .40 --
Comprehension: Oral reading (no paragraph
specific, question 3) .40 --
Comprehension: Oral reading (paragraph 4,
list) .40 --
Comprehension: Listening (paragraph 4,
list) .40 --
Comprehension: Listening (paragraph 4,
question 1) .40 --
Vocabulary concepts: "Simple" (in
isolation) .40 --
Vocabulary concepts: "Simple" (student
chunk) .40 --

Word recognition: "Sammy" (student chunk) .40 --

102

TABLE 6.--Continued

 

SELECTED CUES CASE 7 CASE 8
(221 cues (321 cues
total) total)

Comprehension: Silent reading (oral
questions & answers,

paragraph 1, list) -- .40
Comprehension: Oral reading (paragraph
1 question 3) -- .40
Comprehension: Oral reading (paragraph 1,
question 7) -- .40
Vocabulary concepts: "Famous" (in
isolation) -- .40
Vocabulary concepts: "Dosens" (in
isolation) -- .40
Word recognition: "During" (in isolation) -- .40
Word recognition: "Detective" (in
isolation -- .40
Word recognition: "Picnic" (in isolation) -- .40
Word recognition: "Plates" (in isolation) -- .40
Word recognition: "Detective" (teacher
chunk) -- .40

 

The first column in the table lists the cues collected
by at least 40 percent of the teachers for one or both cases.
The second and third columns show the percentage of teachers
who collected these cues. For example, "Cultural-Home
Background" was collected by 90 percent of the teachers,
for both Cases 7 and 8. "Word recognition: 'Detective'",
on the other hand, was collected by only 40 percent of the
teachers, for Case 8 (Background information cues were the
only ones selected for both cases; that is, all other cues
were slightly different from case to case. The teachers
could not, evidently, collect the exact same one for both

cases.)

103

As Table 6 shows, the proportional agreement for
cues collected was much higher than for the diagnostic
judgments (Table 5); teachers collected the same cues
more often than they arrived at the same diagnostic judgments
for a given case. The Case Inventory may have been partially
responsible for this result; in providing a specific set of
cues it might have increased the likelihood that the same
cues would be collected. The kind of organization it offered
was not available to the teachers when they made their
diagnostic judgments.

As the table shows, background data about the child
was the most frequently requested information, across both
cases, and cultural-home data was the type of background
information most commonly asked for. This might indicate
that the teachers depended on socio-cultural data to a
great degree in attempting to make decisions about the
cases' reading problems. It may further suggest that when
the teachers dealt with reading problems, they did not
necessarily operate from models which were not directly
related to the reading process (e.g., socio-cultural models).

Figure 2 denotes the percentage of all cues collected
by the 10 teachers for Case 7 and 8. The verticle line
indicates the percentage of all cues cOllected, and the
horizontal line the number of teachers who collected these
percentages of cues.

The figure shows limited agreement among the teachers
in their use of the 221 cues collected for Case 7 and 321

collected for Case 8. For example, 55 percent of the total

100 .
95 -
9o 1
85 .
so .
75
7o .
65 -
60 I
55 4
so .
45 «
no
35 I
30 .
25 ,
zo .

Percent of cues collected

15‘
10I

 

104

. case 7 (N=221 cues) mean = 1.90

x ----- x case 8 (N=321 cues) mean = 1.61

 

 

1 2 3 4 5 6 7 8 9 10

Number of teachers collecting cues

Figure 2.--Percent of all cues collected for Cases 7 and 8

by 10 teachers

105

cues collected for Case 7 were collected by only one
teacher, while only 5 percent of the total were collected
by nine teachers. Similarly, nearly two-thirds of all
cues collected for Case 8 were collected by one teacher,
and only 5 percent of the total were collected by nine.

In summary, the data presented in Figure 2 show
that there was little agreement in cue collection among the

10 teachers for either case.

Commonality score.--commonality scores were cal-

 

culated to determine the degree of agreement between each
teacher's diagnoses and the group of diagnoses made by all
other teachers for the same case. The score takes into
account which diagnostic judgments, and how many, were made
by each teacher. The same analyses were performed with cues
collected. Table 7 shows the diagnostic commonality scores
for Cases 7 and 8.

The first column in the table gives the representative
values of the Phi correlations; the second column indicates
the frequency count for each of the values. For example,
two teachers for Case 7, and three for Case 8, correspond to
the .4 value. The distributions were quite normal for both
cases.

The table shows that the mean commonality scores for
diagnostic judgments made by all 10 teachers were .45 and
.45, for Case 7 and 8, respectively. This means that, on

the average, each teacher's diagnosis for a given case

106

TABLE 7.--Diagnostic commonality score for Cases 7

and 8 and 10 teachers

 

HISTOGRAM/FREQUENCES CASES = 1-10
MIDPOINT FOR 6.V6 (Each X=1)

-1.0000
-.90000
-.80000
-.70000
-.60000
-.50000
-.40000
-.30000
-.20000
-.10000
0.0
.10000
.20000
.30000 +
.40000 +XX000
.50000 +X000
.60000 +
.70000 +Xx0
.80000 +x0
{.90000 +X
1.0000 +

X=case 7 mean=.45
N=10 standard dev.=.29

0=case 8 mean=.45
N=10 standard dev.=.23

+-;-++-+-+4—+-+4—+-+
O

+
N
N
O

l . \ I

I

TOTAL 10 (INTERVAL WIDTH=.10000)

 

included about half of the most frequently-mentioned
diagnostic judgments made by the other teachers.

Table 8 gives the cue commonality scores for Cases 7
and 8.

The first column in the table shows the representative
value for the Phi correlations; the second column indicates

the frequency count for each of the values. For example,

107

TABLE 8.--Cue commonality scores for Cases 7 and 8

and 10 teachers

 

HISTOGRAM/FREQUENCIES CASES = 1-10

MIDPOINT FOR 6.V6 (EACH X=1)
-1,0000 + X=case 7 mean=.56
-.90000 + N=10 standard dev.=.ll
-.80000 +
-.70000 + 0=case 8 mean=.55
-.60000 + N=10 standard dev.=.l4
-.50000 +
-.40000 +
-.30000 +
-.20000 +
-.10000 +
0.0 +

.10000 +

.20000 +0

.30000 +X

.40000 +0

.50000 +XXX00

.60000 +XXXX00

.70000 +XX0000

.80000 +

.90000 +

1.0000 +
TOTAL (INTERVAL WIDTH=.10000)

 

three teachers, for Case 7, and two, for Case 8,
corresponds to the .5 value.

As this table reveals, the mean cue commonality scores
for the two cases were higher than the diagnostic commonality
scores (Table 7). This suggests that, on the average, the
teachers agreed more on what cues to collect than on what

diagnostic judgments to make.

108

Inter-clinician agreement correlations--the inter-
clinician correlations denoted the agreement between one
teacher and each of the others on diagnostic judgments and
cues collected for five cases (Phi correlations are the
Pearson Product Moment correlations with the assumption that
the two variables are dichotomos and not necessarily normally
distributed). Specifically, these correlations indicated
the degree to which particular cues and diagnostic judgments
in one teacher's diagnosis were present in or absent from,
another teacher's diagnosis. Table 9 shows the inter-
clinician correlations for diagnostic judgments for Cases
7 and 8.

The first column in the table shows the representative
value of the Phi correlations; the second column gives the
frequency count for each of the values. For example, the
frequency count for the .2 value was 3, for Case 7, and 2,
for Case 8.

The data in Table 9 indicate that the agreement
scores for diagnostic judgments were near zero, for both
cases. In other words, on the average, the teachers showed
very limited agreement (if any) on their diagnostic judgments.

Table 10 shows the inter-clinician correlations for
cues collected for the two cases.

The first column in the table gives the representative
Phi coefficient values, the second column the frequency count
for each of the values. For example, the frequency count for

the .2 value were 6, for Case 7, and 3, for Case 8.

109

TABLE 9.--Inter-clinician correlations for diagnostic

judgments for Cases 7 and 8 and 10 teachers

 

HISTOGRAM/FREQUENCIES CASES = 1-45
MIDPOINT FOR 7.V7 (EACH X=1)

-1.0000
-.90000
-.80000
-.70000
-.60000
-.50000
-.40000
-.30000
-.20000 +XXXXXXXXXXXXX00000000
-.10000 +XXXXXXXXXXXXX000000000000000000
0.0 +XXXXX000000

.10000 +XXXXXXXXXX00000000000

.20000 +XXX00

.30000 +X

.40000

.50000

.60000

.70000

.80000

.90000
1.0000

X=Case 7 mean= -.04
N=45 standard dev.= .13

0=Case 8 mean= -.03
N=45 standard dev.= .11

+-++-+-++-+-+

+-++-+-++-+

TOTAL 45 (INTERVAL WIDTH= .10000)

 

As Table 10 indicates, the inter-teacher agreement
scores for cues collected was around zero, for both cases.
This suggests that, on the average, the teachers showed

little agreement in collecting cues for a given case.

110

TABLE lO.--Inter-c1inician correlations for cues collected

for Cases 7 and 8 and 10 teachers

 

HISTOGRAM/FREQUENCIES CASES = 1-45
MIDPOINT FOR 7.V7 (EACH X=1)

-1.0000
-.90000
-.80000
-.70000
-.60000
-.50000
-.40000
-.30000
-.20000 +X
-.10000 +XXXXXXXXX00000000
0.0 +XXXXXXXXXXXXXXXX000000000000000000000000
.10000 +XXXXXXXXXXXX0000000000
.20000 +XXXXXX000
.30000 +X
.40000
.50000
.60000
.70000
.80000
.90000
1.0000

X=Case 7 mean= .03
N=45 standard dev.= .ll

0=Case 8 mean= .01
N=45 standard dev.=.07

+-++-+-++-+-+

+-+4-+-++-+

TOTAL 45 (INTERVAL WIDTH= .10000)

 

Product measures: Summary

 

The product measures presented in this section
suggest the following:
1. Proportional agreement among teachers on both diagnostic
judgments made and cues collected was very limited.
2. The most frequently-mentioned diagnostic judgments
differed from case to case. This seems reasonable,

given that the two cases included different reading

111

problems and different cues.

3. The most frequently-mentioned cues for both cases were

those providing background information. This suggests
'that teachers may apply models of students' personal
characteristics to explain students' reading
achievement. Many of the models teachers use, do not
seem to deal directly with reading or reading skills;
rather, they center around the socio-psychological
aspects of reading behavior. A possible explanation
for this observation is that teachers usually take
many classes in the socio—psychological areas, without
focusing on how they affect reading or reading
acquisition.

4. Inter-clinician correlations for both diagnostic
judgments and cues collected were near zero, meaning
that the teachers showed little agreement on diagnostic
judgments or on cues collected for a given case.

Results across the two cases were almost the same,
suggesting that the teachers employed similar diagnostic
processes for both cases.

6. On all three measures, agreement on cues collected was
slightly higher than on diagnostic judgments made. That
is, the teachers tended to show closer agreement in
collecting cues than in making final diagnostic judgments.

It must be noted that the results are based on data

obtained from a very small number of teachers. But it should

112

also be noted, however, that the findings are highly
consistent across the two cases, and that similar

studies, involving reading specialists and learning
disability personnel, have provided similar results

(Van Roekel, 1979; Vinsonhaler, 1979). Even so, replication
studies with classroom teachers are needed not only to check
the validity of findings from this investigation, but to
determine whether simulated cases accurately represent real

classroom cases of reading difficulty.

Part 2: Process measures

 

A question of interest to the study of clinical
problem-solving behavior is how teachers gather information
about a case and reach final diagnoses. Specific elements
of problem-solving behavior which can be measured and com-
pared include length of interaction, number of cues collected
on a case, number of final stated diagnoses for a case, and
the relationships between cues collected and teachers'
verbatim comments about the cues. The following illustration
of a process of diagnosis intends to clarify the way
teachers' memories and strategies affect these specific
elements of the problem solving behavior.

As noted at the beginning of this chapter, different
teachers may Operate from different memories and
strategies. A teacher who operates from a skill-model
orientation, for example, will tend to collect and process

information in a manner different from that of a teacher

113

Operating from a socio-pyschological orientation. The first
teacher may be an inductive reasoner, the second a

deductive reasoner. Differences in memories and strategies,
the reader will recall, may be attributed, in part, to
different kinds of training teachers received.

Following is an example of the diagnostic process
followed by a teacher who Operates from a socio-psychological
orientation. This example is based on the behavior of one
of the teachers (no. 7) who participated in this study.

After receiving initial contact information about the
case, the teacher asked for information pertaining to the
child's personality. She explained: "I feel children's
personality has a lot to do with how they perceive them-
selves. It has an effect on classwork". The teacher
listened to a tape of the child reading a passage and
observed that "she was a very poor reader and made no
effort to sound words". The teacher then asked for
information about the child's cultural-home background, and
explained: "I wanted to see if she was Black. Accented
like a Black. Was surprised she was Spanish. My hunch:
she's under stress. Might be emotional/social problem, not
an intellectual one". The next cue the teacher requested
was an audio tape on which the child read the words
"Pittsburgh" and "Pennsylvania" (this tested the child's
auditory-sound blending). The teacher commented: "She

didn't seem to be able to sound out and blend. (However),

114

sometimes English-as-a-second-language-kids have this
problem. (So), it might be a waste of time to do phonics --
they don't have the same sounds. Silly to try and get them
to do it."

At this point, having collected four cues, the teacher
was not sure what else to ask for. She decided to look at
Test Results and "was very surprised at how well (the child)
did on the tests." The teacher collected two more cues
(written questions and answers for silent reading, and
the word ”picnic" in student sentence) after which she
said: "I'm getting swamped here with all this information.
I'm surprised she did as well on tests as she did. Most of
her problem might be emotional/social." The teacher asked
for one last one -- the eighth (comprehension question
following a silent reading) and said: "Some children can
read silently and comprehend -- that's O.K." The teacher
concluded the session with two stated diagnostic judgments
about the case: (1) that emotional or social problems
might have been hampering the student's school work, and
(2) that she was insecure about herself.

This example of a process of diagnosis illustrated
some of the elements of teachers' diagnostic processes which
can be measured and compared. As mentioned at the beginning
of this section, among those elements are length of
interaction with a case, number of cues collected, and

number of final stated diagnoses made by the teachers.

115

Teachers' performance on the two cases

Teachers' performance in terms of length of
interaction, number of cues collected, and number of
diagnostic judgments:

Examining teachers' performances in terms of these
variables is one way the researcher can explore some of the
characteristics of the teachers' clinical problem-solving
behavior. Analyses of process measures for this study
were based on data obtained during both the observational

and the debriefing sessions.

Length of interaction

 

Table 11 shows the length of time the teachers
interacted with Cases 7 and 8.

The first column in the table represents the length
of the interaction, in minutes; the second column shows the
frequency count for each of the time variables. The table
indicates, for example, that two teachers interacted with
Case 7 for 40 minutes, and two interacted with Case 8 for
40 minutes (See Appendix G for raw data on measures of
time, cues collected, and diagnostic judgments for the 10
teachers on the two cases.)

As the table shows, different teachers interacted
with the cases for different amounts of time. Length of
interaction ranged from a minimum of about 25 minutes to a

maximum of 60 minutes (arbitrary cut-off time), with a mean

116

TABLE ll.--Length of interaction for Cases 7

and 8 and 10 teachers

 

HISTOGRAM/FREQUENCIES CASE = 1-10

MIDPOINT FOR 1. TIME (EACH X= 1)

0. +

5.0000 + X= Case 7 mean= 41.3
10.000 + N=10

15.000 +

20.000 + 0= Case 8 mean= 48.1
25.000 +XXO N=10

30.000 +

35.000 +XX

40.000 +XX00

45.000 +000

50.000 +XX

55.000 +X0

60.000 +X000

TOTAL 10 (INTERVAL WIDTH= 5.0000)

 

time of 41.3 minutes for Case 7, and 48.1 minutes, for Case
8. This fihding suggests that, as with most other psycho-
logical variables (which were investigated in various
studies), teachers' interactions with simulated cases

reveal marked individual differences.

Number of cues collected

 

Similar individual differences were found in the
number of cues the teachers collected. Table 12 displays
this information.

The first column in the table gives the number of

cues collected; the second the frequency count (i.e., number

117

TABLE 12.--Number of cues collected on Cases 7

and 8 and 10 teachers

 

HISTOGRAM/FREQUENCIES CASES=CASE 1-10

MIDPOINT FOR 1.CUE7 (EACH X= 1)

0. + X= Case 7 mean= 42.3
10.000 +0 N=10

20.000 +X

30.000 +X0 = case 8 mean= 51.4
40.000 +XXXXXO N=10

50.000 +XX00

60.000 +00

70.000 +00

80.000 +0

90.000 +X

100.00 +

TOTAL 10 (INTERVAL WIDTH= 10.000)

 

of teachers) for each of these cue values. For example, one
teacher collected about 30 cues for Case 7, and one collected
30 for Case 8.

As the table shows, different teachers collected
different numbers of cues, from a minimum of about 10 to a
maximum of about 90. In other words, the teachers tended to
collect different amounts of information to reach diagnostic
judgments.

These differences might have stemmed from the
different models the teachers used (if they used any).

A teacher operating from a comprehensive model was likely
to gather more information about the case than one operating

from a non-comprehensive model. Indeed, teacher No. 7,

118

who seemed to operate from a socio-psychological (non-
comprehensive) model, collected the fewest cues of all
teachers (17 for Case 7; 12 for Case 8), and made the

least number of diagnostic judgments (3 for Case 7, 2 for
Case 8). Her diagnostic judgments for Case 8 dealt only
with emotional problems that might have affected the child's
reading performance (It is interesting to note that, during
the observational session, this teacher wanted to stOp
gathering information about the case after having collected
only 10 cues. She explained: "I think I know what the

problem is; I already made up my mind").

Number of diagnostic judgments

 

Table 13 shows the number of diagnostic judgments the
teachers made for Cases 7 and 8. The first column
represents the number of diagnostic judgments made; the
second column shows the frequency count for each of these
values. For example, two teachers made four judgments for
Case 7, and one made four diagnostic judgments for Case 8.

As Table 13 indicates, the teachers differed in the
number of diagnostic judgments they made for a given case.
The number of judgments ranged from 2 (made by one teacher
for Case 8) to 14 (made by two teachers for Case 8). As
in the case of cues collected, these differences might have
been caused by the different models of reading and

diagnosis from which the teachers operated.

119

TABLE 13.--Number of diagnostic judgments for Cases 7 and

8 made by 10 teachers

 

HISTOGRAM/FREQUENCES CASES = 1-10

MIDPOINT FOR l.DIAGNOSE (EACH X= l)
0. + X= Case 7 mean= 7.2
2.0000 +0 N=10

4.0000 +XXO

6.0000 +XX 0: Case 8 mean= 9.0
8.0000 +XXXO N=10

10.000 +XX000

12.000 +X00

14.000 +00

16.000 +

18.000 +

20.000 +

TOTAL 10 (INTERVAL WIDTH= 8.0000)

 

As with length of interaction and number of cues
collected, Table 13 also suggests that similar individual
differences existed in teachers in terms of the number of

diagnostic judgments they made for a case.

Summary

The data presented in Table 11 through 13 suggest
that the teachers differed significantly from one another in
the length of time they interacted with the simulated cases,
the number of cues they collected, and the number of diagnostic

judgments they made.

120

Since differences among the teachers were observed
for each of these variables, it might be worth asking what
relationships existed among these variables. Table 14
summarizes the mean correlations among length of interaction,
number of cues collected, and number of diagnostic judgments

for Cases 7 and 8.

TABLE l4.--Correlations among length of interaction, number
of cues collected and number of diagnostic

judgments for Cases 7 and 8 and 10 teachers

 

TIME 1.0000
(7)CUES .4942 1.0000

DIAGNOSES .4650 .6109 1.0000

 

TIME .4970 .5120 .4574 1.0000
(8)CUES .5172 .4679 .1450 .3278 1.0000
DIAGNOSES .5381 .5390 .9672 .4614 .2421 1.0000
TIME CUES DIAGNOSES TIME CUES DIAGNOSES
CASE 7 CASE 8

 

 

Because Of the small number of teachers involved, no
generalizations can be drawn from these data. Descriptive
statistics, however, suggest that, in this study, the
following observed relationships existed:

1. Between length of interaction with a case and number of

cues collected (.49 for Case 7, .32 for Case 8). On

121

the average, the more time teachers interacted with a
case, the more cues they collected.

Between length of interaction and number of diagnostic
judgments made (.46 for Case 7 and Case 8, respectively).
On the average, the longer teachers interacted with a
case, the more diagnostic judgments they made.

Between number of cues collected and number of
diagnostic judgments made (.61 for Case 7, .24 for

Case 8). In general, the more cues teachers collected
for a given case, the more diagnostic judgments they
made. (The marked differences in mean values between
Cases 7 and 8 are probably due to differences in the
number and kinds of cues provided. Case 8 included more
items of information and more varied types of information
than Case 7. Thus, to attempt a diagnosis on Case 8,

the teachers had to collect more cues than they did for
Case 7.)

Between length of interaction with Case 7 and length of
interaction with Case 8 (.49). Most teachers interacted
with the two cases for about the same length of time.
Between total number of cues collected for Case 7 and
total collected for Case 8 (.46). Teachers who collected
many cues for one case tended to collect many for the
other as well; conversely, teachers who collected few
cues for one case also tended to collect few for the
other.

These differences among teachers in the number of

122

cues they collected might be explained by the Agreement
Corollary. It may well be that different memories or models
from which the teachers operated prompted them to collect
different amounts of information. A skill-oriented teacher,
for example, might have felt it necessary to collect a large
amount of data to check the specific skills the child lacked.
A teacher operating from a socio- psychological model, on
the other hand, might have based his/her diagnosis on few
cues, gathered in the narrow area of psychological and
social background.
Differences in the teachers' personal characteristics

(such as self-confidence, independence/dependence, etc.)
might also have accounted for the differences in their
diagnostic performances. Some peOple will always collect a
large amount of information on which to base a decision,
while others will use relatively little data.
6. Between number of diagnostic judgments made for one

case and number of diagnostic judgments made for

another (.96). Teachers tended to make approximately

the same number of judgments for each case, relative

to each other.

The explanation for this observation might also

lie in the Agreement Corollary. The memories and models
from which teachers operate affect the number of judgments
they make. A teacher operating from a psycho-social model,

for example, will tend to entertain a limited number of

123

diagnoses about a case. The same is true of a teacher who
operates from a skill-model orientation. Consistent use of
one distinct model probably leads the teachers to make
similar diagnoses for different cases of reading difficulties.
It is logical that, if the teachers make similar cross-case
diagnoses, they entertain about the same number of diagnostic
judgments.

The data presented in this section suggest that,
although teachers differed from one another in their
diagnostic performances on the three variables, they were
fairly consistent in their own interactions with the two
cases of reading difficulty.

Analysis of the relationships between cues collected and
teachersTiverbatim comments about these cues

 

While interacting with the simulated cases, the
teachers were asked to verbalize their thoughts about the
information they collected. They were asked to do this
again during the debriefing session. Their verbatim
statements were audiotaped and transcribed by a clinical
observer. Data analysis explored the relationships between
cues collected and the teachers' comments about the cues.

The examination of these relationships was guided
by a basic question: What diagnostic strategies did the
10 teachers employ? This question was approached in two
ways: (1) through analysis of quantitative features of

teachers' diagnostic processes, and (2) through analysis

124

of the types of verbatim comments the teachers made about
the cues.

The data selected for analysis were those collected
during the teachers' initial diagnostic interactions (five
with Case 7 and five with Case 8). The researcher felt
that the experience of interacting with the first simulated
case might have altered the manner in which teachers dealt
with the second case (session 1 served as a training session
for session 2), and he preferred to focus on the teachers'
initial processes of gathering and evaluating information
about the reading difficulties.

Table 15 lists three of the quantitative features of
the teachers' diagnostic processes -- (1) number of verbatim
comments made, (2) average cue percent time (i.e., proportion
of time elapsed when half of cues were collected), and (3)
average verbatim percent time (i.e., proportion of time
elapsed when half of verbatim comments were made).

As Table 15 suggests, marked individual differences
existed among the teachers in relation to number of
verbatim comments they made about the cues they gathered
and the case. They could have made from zero to l or more
comments about each of the cues.

The table also indicates that, for both cases,
teachers tended to collect cues throughout the interaction.
They also tended to talk and express their thoughts about
the case in relation to the cues collected throughout

the interaction. As the table shows, on the average, about

125

TABLE lS.--Number of verbatim comments, mean cue percent
time and verbatim percent time for Cases

7 and 8 and 5 teachers

 

STATISTIC TEACHERS
*1 2 3 4 5 6 7 8 9 10 X
Number of verbatim
comments 59 82 60 78 85 72.8
(7)

 

Average cue
percent time .535 .637 .513 .635 .625 .589

 

Average verbatim
percent time .470 .565 .402 .521 .484 .488

 

 

Number of verbatim
comments 50 41 64 34 69 50.6

 

Average cue
percent time .513 .655 .465 .630 .493 .511

 

(8) Average verbatim
percent time .518 .578 .471 .565 .509 .528

 

 

 

half of the cues were collected after 60 percent (.589) of
the session with Case 7, and after 55 percent (.551) of the
session with Case 8. This means that a somewhat larger
number of cues was collected during the second half of the
observational sessions than during the first half. This may
suggest that, as the teachers felt time was running short,
they accelerated the rate at which they gathered information

about the case.

126

The teachers' cognitive or diagnostic processes
were also explored through analysis of the types of verbatim
comments they made about the cues they gathered during the
interactions and the debriefing sessions that followed.

This anlaysis provided data about some of the charac—

teristics of the teachers' clinical problem-solving behavior,

including: (1) how much they observed specific reading
behaviors, and (2) whether they posed and tested hypotheses
about the cases' problems. Figure 3 shows the percentage

of verbatim comments and their relationships to cues.

This figure shows the distribution of teachers'
verbatim comments among 10 different types of possible
relationships between cues and statements. The verticle
line represents the percentage of verbatim comments made
by all 10 teachers for Cases 7 and 8. The horizontal line
denotes the types of relationships which existed between cues
and statements. (See pages 86-87 for description of each of
these relations.

The data indicates that:

l. The teachers did not formulate hypotheses about the
reading difficulties of a given case. That is, their
data-gathering was not oriented toward confirmation
or disconfirmation of hypotheses about the case's
problems. Only seven percent of the teachers' total
verbatim comments dealt with hypotheses.

2. The teachers tended to observe rather than assess or

evaluate students' reading difficuties. Twenty-four

Percent of verbatim comments made by

10 teachers for Cases 7 & 8

127

V

100 1
95'
90-
85.
80.

N=617

28X
20 1 271

242

 

. F'—‘—1'
5 11 """"‘"'“ 1% 62

6E 0.0: 4: . 32
V r V I f I W I
l 2 3 4 5 6 7 8 9 10
confirm. neither. generali- strategy remedia-
hunch zation tion
disconfirm specific other no relation philosophy
hunch. comment to one of reading

I 7x I 33: on 10:

 

 

 

 

 

 

 

 

 

 

‘

]

statements about statements about other
hypotheses cues

Type of verbatim comments

Figure 3.--Cue to statements relationships

128

percent of the statements related specifically to the
cue collected (relation type 4). Although the teachers
did relate several observations to one another
(relation type 5 - twenty-eight percent), they did not
go beyond the mere observation level (to a diagnostic
level).

The teachers tended to talk about what they did as they
interacted with a case (twenty-seven percent of their
statements self-described the way in which they
gathered information). They did not, however,

describe how they reached diagnostic judgments.

Process measures: Summary

 

The process measures presented in this section

revealed the following:

1.

The teachers differed significantly in terms of how long
they interacted with the simulated cases, the number of
cues they collected, and the number of diagnostic
judgments they made.

There were observed relationships among length of
interaction, number of cues collected, and number of
diagnostic judgments. The relationships held for
interactions with both cases.

The teachers were inferential problem solvers, to the
degree that they problem solved at all. They did not
pose and test hypotheses about the cases, and did not

gather information to confirm or disconfirm hunches.

129

Rather, they gathered data and commented about it. They
were observers of children's behaviors, but they did not
process the observed information in a manner that led to

accurate and complete diagnostic judgments.

Part 3: Classroom interviews

 

The main purpose of the classroom interviews was
to explore teachers' diagnostic skills in the natural
classroom environment and to investigate the relationships
between these diagnostic skills and those exhibited in the
laboratory situations.

The interviews were intended to help provide
answers to two questions: (1) Is the examination of
teachers' interactions with simulated cases in Laboratory
situations a valid means of studying teachers' diagnostic
skills? and (2) What are some of the diagnostic processes
employed by teachers in the natural classroom environment?

This part of the investigation was approached in
two ways: (1) through analysis of the diagnostic categories
most frequently mentioned by teachers in the classroom,
and (2) through analysis of the similarities and
differences between diagnostic categories mentioned in the
Laboratory and the classroom situations.

Following a discussion of the results of these
analyses, several other findings, based on a few other

interview questions, are discussed.

130

What are the diagnostic categories most frequently
mentioned by teachers in the classroom?

 

During the classroom interviews, the teachers
discussed various elements of reading to which they
attend when reaching diagnostic decisions about students'
reading performances. A domain of these diagnostic
categories was created, and redundancies were eliminated.

The final list contained six specific categories, grouped
into 6 general classes: Comprehension, Vocabulary, Word
recognition, Word attack, Oral reading and Student's
characteristics (See Appendix H: Diagnostic domain based
on classroom interviews).

A question of interest was, what proportion of
teachers mentioned each category. This information is
important because it might give some idea of which diagnostic
categories teachers most frequently deal with, or think of.
Table 16 shows the diagnostic categories most frequently
mentioned in the classroom interviews.

The first column in the table lists the diagnostic
categories mentioned by at least 30 percent of the teachers.
The second column shows the exact percentage of teachers
who mentioned each category. For example, "Sight Vocabulary"
was mentioned by 100 percent of the teachers (n=10), whereas
"Blends” was mentioned by only 30 percent (n=3).

A close look at these data about the diagnostic

categories reveals that:

131

TABLE l6.--Twenty—eight most frequently mentioned diagnostic

categories in the classroom (Total of 96)

 

PROPORTION OF
TEACHERS MEN-
TIONING THE

 

DIAGNOSTIC CATEGORIES CATEGORIES
Sight vocabulary 1.00
Comprehension .90
Silent reading comprehension .80
Oral reading .80
Oral reading comprehension .70
Word recognition .70
Vocabulary .60
Endings .60
Context .60
Listening comprehension .50
Word analysis .50
Decoding .50
Phonetic analysis .50
Beginning sounds .50
Short vowels .50
Word attack .40
Chunking .40
Sounds .40
Word family approach .40
Able to sound out .40
Phonetic skills .40
Print translation .40
Substitutions .40
Factual and inferential comprehension .30
Vocabulary meaning .30
Guesses .30
Ending Sounds .30

Blends .30

 

132

1. Five of the six general categories composing the total
diagnostic domain (Comprehension, Vocabulary, Word
recognition, Word attack, and Oral reading) were
mentioned by at least three teachers; however,
different teachers discussed different sub-skills
under each of these categories.

2. Those categories mentioned by at least 70 percent
of the teachers were very general: Sight vocabulary,
Comprehension, Silent reading comprehension, Oral
reading, Oral reading comprehension and Word
recognition, This suggests that the teachers rely
on the same general diagnostic categories in reaching
decisions. The experimental design might have been
partially responsible for this finding. All ten
teachers were given the same descriptions of reading
difficulties, and these descriptions may have
cued them to mention similar general diagnostic
categories. The teachers seemed to differ, however,
from one another, on the sub-skills within each of the
general categories (e.g., miscues, skip words,
punctuation, etc.).

Figure 4 shows the percentage of total diagnostic
categories (n=96) mentioned by the teacehrs in the classroom
interviews.

The verticle line gives the percentage of diagnostic
categories mentioned, and the horizontal line shows the

number of teachers who mentioned the categories. For example,

133

N = 96
mean = 2.32

75.

654
501
554
50.
45.
404
35.
30.
25.
20
15
10

Percent of diagnostic categories mentioned
\I
1?

5.-

v

 

““‘fi--t$

I T

9 10

I I I r I I

l 2 3 4 5 6

‘34!)

9°40

Number of teachers mentioning category

Figure 4.--Percent of the diagnostic categories mentioned
by teachers in the classroom

134

52 percent of the categories were mentioned by only one
teacher, and 1 percent were mentioned by nine and ten
teachers, respectively.

It should be noted, however, that teachers' agreement
on diagnostic categories mentioned in the classroom interviews
is slightly higher than their diagnostic agreement for Cases
7 and 8 (Figure 1). This is so because, during the
interviews, the teachers tended to make general statements,
whereas during the laboratory interactions with the cases
they were more case-specific.

It must be pointed out that the classroom interviews
were based on four descriptions of deficient readers which
were given to the teachers. The teachers were asked to
match two of these descriptions with students in their own
classrooms. As one might expect, the teachers found
differences among the described readers, and made different
diagnostic judgments about them. The number of teachers who
matched readers in their own classrooms with the descriptions
varied from seven (70 percent), for the description of
"Michelle", to three, for the description of "Diane" (Appendix
E lists the number of teachers who noted similarities between
children in the descriptions and children in their own

classrooms).

135

Similarities and differences between teachers' diagnostic
categories In laBOratory and classroom situations

 

After the domain of diagnostic categories (judgments)

mentioned by the teachers in the classroom was compiled, it

was compared with the domain of judgments made by the teachers

in the laboratory setting. This was done for two reasons:

1.

It was hOped the comparison might point to the existence
or absence of a model of reading held by the teachers, or
at least, to a consistency or inconsistency in their
thinking about reading problems. If similar diagnostic
categories are mentioned by teachers in both laboratory
and classroom situations, it may suggest that they operate
from a model, or from a certain"memory" which contains
these cateOgires.

Checking for similarities and differences between
diagnostic categories mentioned in the laboratory
situations and those mentioned in the classroom was one
way of exploring whether carefully-desi

observational studies in laboratory settings are

reliable and valid means of investigating the clinical
problem-solving behavior of teachers. Similarities

among the categories, it was thought, would suggest

that teachers perform in both situations in a similar
manner, or at least from the same model or with the

same memory.

136

A careful look at the data reveals that, on the

average, 50 percent of the diagnostic categories mentioned
by the teachers in laboratory settings were also mentioned

during classroom interviews (See Appendix F).

Table 17 shows the numbers of diagnostic categories

mentioned by each teacher in laboratory setting (for the two

simulated cases) and during classroom interviews.

TABLE l7.--Number of diagnostic categories mentioned by

teachers in laboratory and classroom situations

 

 

 

 

 

CONDITION TEACHERS
l 2 3 4 5 6 7 8 9 10 X
Laboratory:

Cases 7 & 8 13 15 24 21 24 6 5 l9 17 18 .16.2
Classroom 30 18 22 24 19 26 22 26 17 26 23.0
Number of
categories
mentioned
in both 8 9 13 13 6 5 3 8 7 11 8.5
laboratory
and class-
room

 

 

 

 

 

 

 

 

 

 

 

As the table indicates, the number of diagnostic

categories mentioned by the teachers in the classroom

outnumbered the number of diagnoses made in the laboratory.

This may suggest that the domain of diagnostic categories

simulated by the two simulated cases was more limited than

 

137

that which evolved from the natural classroom setting.

Looking carefully at the kinds of diagnostic statements

mentioned in both the laboratory and the classroom

(Appendix F), we can see that the statements tended to

fall into two categories:

1.

general statements, such as: "Vocabulary concepts:
knows major concepts"; "Sight vocabulary: good". In
other words, these are general statements about the
reading act which do not require careful and detailed
diagnosis of a child's reading difficulties;

statements which seem to be characteristic of individual
teachers rather than individual students, such as:
"Insecure about herself"; "the manifestations of her
problems are dialectical based". In other words, some
statements about the reading act seem to emerge
primarily from the teacher's philosophical orientation
(or model) about the causes of reading difficulty,
rather than from some specific aspect of the child's
reading performance. (An extreme example was the

set of statements made by teacher no. 7, who, in both
laboratory and classroom situations, emphasized, above
all, that children's emotional problems and self—concept
are the main causes of reading deficiency. At the
opposite extreme was teacher no. 3, who, in both
settings, conducted thorough and detailed diagnoses and
considered many variables as potential causes of the

reading difficulty).

138

Diagnostic judgments mentioned in either laboratOEy
or classroom situatIOn Tbut not in both)

 

Diagnostic categories mentioned in the laboratory
but not during classroom interviews (see Appendix F) tended
to be very case specific. They applied directly to the
simulated case the teacher was attempting to diagnose, and
frequently consisted of some sub-skills of the reading act
applicable to this case (e.g., "punctuation: does not
pay attention to";"syllabication skills: poor"; "plural
endings: poor," etc.). During classroom interviews, when
the teachers discussed either the individual students in
their classroom or their diagnostic practices in general,
specific sub-skills, which were case specific, were not
mentioned.

Diagnostic statements mentioned in the classroom
but not in the laboratory settings (See Appendix H)
tended to fall into two categories:

1. Case specific statements, i.e., statements related
directly to the individual student whom the teacher was
describing during the interview. Especially prevalent
were statements which related to the child's factors.

2. General statements, usually mentioned by the teacher as
she described her instructional and diagnostic
practices in reading (e.g., syllables, compound words,
decoding). These were not mentioned in the laboratory
setting because the interactions with simulated cases
forced the teachers to be as specific as possible in

their diagnoses.

139

The data presented in Table 18 show the diagnostic

categories most frequently mentioned in the laboratory and

classroom situations.

TABLE 18.--Most frequently mentioned diagnostic categories

in laboratory and classroom settings

 

 

 

 

DIAGNOSTIC CATEGORIES LABORATORY CLASSROOM
Case 7 Case 8 Children in
specific specific general
Sight vocabulary 5 7 (12) 10
Comprehension 2 4 ( 6) 9
Oral reading 4 3 ( 7) 8
Silent reading comprehension 0 0 ( 0) 8
Oral reading comprehension l 2 ( 3) 7
Word recognition 5 l ( 6) 7
Endings 5 0 ( 5) 6
Context 2 l ( 3) 6
VOcabulary 3 3 ( 6) 6
Beginning sounds l 2 ( 3) 5
Short vowels 0 1 ( l) 5
Listening comprehension 0 2 ( 2) 5
Decoding 0 0 ( 0) 5
Print translation 0 3 ( 3) 4
Phonetic skills 2 3 ( 5) 4
Word attack 2 3 ( 5) 4
Chunking 2 l ( 3) 4
Word family approach 0 0 ( 0) 4
Factual & inferential comp. l 2 ( 3) 3
Vocabulary meaning 2 3 ( 5) 3
Guesses 2 2 ( 4) 3
Ending sounds 0 l ( l) 3
Blends 0 2 ( 2) 3

 

 

These data indicate that, in general, the diagnostic

categories mentioned most often in the laboratory were also

mentioned (and usually more frequently) during the classroom

140

interviews. This may suggest that teachers think of these
general reading categories in both laboratory situations
and the natural classroom environment.

The few differences which were noted between the
categories the teachers entertained in the laboratory setting
and in the classroom (i.e., Silent reading comprehension,
Decoding, and Word family approach) might have stemmed from
the fact that the Laboratory interactions were case-specific,
and thus discouraged the teachers from considering the
diagnostic categories they would consider in the classroom.

In summary, looking at similarities and differences
between diagnostic statements mentioned in the laboratory
setting and those mentioned in the classroom, we noted that
the diagnostic categories most frequently mentioned in the
laboratory were also mentioned in the classroom. Those
which were mentioned in only one setting tended to be either
(1) case specific to the simulated case, or to the
individual student described by the teacher in the classroom,
or (2) general comments, usually about the instructional and
diagnostic practices the teachers mentioned during classroom
interviews.

Overall, (when one excludes these two types of
diagnostic categories), the data indicate that teachers
appear to rely on similar diagnostic categories in both
laboratory and classroom settings. This suggests that
whatever the diagnostic "memory" teachers have, it is
dominant in their interactions with cases of reading dif-

ficulties, whether these interactions take place in a

141

laboratory setting or in the classroom.

Other findings

 

As stated (p. 129), the interviews with the
teachers included several additional questions related
to reading diagnosis and remediation;. the teachers'
responses to these questions are discussed below.

An important question about the processes of reading
diagnosis and remediation is whether teachers Offer
differentiated remediation plans for different reading
problems or different children. The question is important
because it deepends the investigation into teachers' memories
and their functional models of reading diagnosis. One can
expect that, the more complete the memories teachers have,
and the more complete and sophisticated their models of
reading are, the more differentiated the remediations they
offer will be. These different remediations will depend on
the specifics of the reading characteristics exhibited by
the different students the teachers deal with. If the
teachers' memories are not intact, and if they either lack
a model of reading or have an incomplete one, they will
probably offer similar (or standardized) remediation plans
for all cases of reading difficulty they encounter.

Following is a list of remedial plans mentioned
by each of the 10 teachers during the classroom interviews.

It is composed of only those remediations which were mentioned

142

for both of the cases whose reading difficulties the

teachers addressed:

Teacher

Teacher

Teacher

Teacher

Teacher

Teacher

Teacher

1:
2:

3:

5:
6:

Work on phonetic skills.

a)

b)

a)

b)

C)

d)

a)

b)

Take words out of context and see how the child
can handle it.

Encourage the child to work.

Work on beginning, middle and ending sounds.
Have the child work with an aid on short-
paragraph stories.

Have the child go to a reading specialist for
half an hour each day.

We'll work on short paragraphs with many
questions.

Work on print translation.

Focus on context until the child realizes

that things make sense.

A lot of practice.

a)
b)

a)

b)

C)

d)

e)

Work on vocabulary.

Spend time with him on an individual basis.
Try to get the child to read words he/she
cannot sound out.

Send words home, ask mother or father to help
the child.

Ask the child to look at the picture.

He/she needs to pay more attention.

Have games with words.

143

Teacher 8: a) Give the child additional material to read.

b) Have him/her read on a one-to-one basis.

c) Have him/her work on worksheets.

Teacher 9: a) Send the child to a reading teacher.

b) Do more individualized work with the child

on a one-to-one basis.

Teacher 10:a) Encourage the child to do more by himself.

b) Encourage the child to ask for help.
c) Send the child to a reading lab.

An examination of this list of remediations

reveals the following:

1.

Most of the recommended remediations are very general,
such as: "Work on print translation"; "work on phonetic
skills", etc.

Many of these remediations call for additional practice
rather than dealing with the specifics of the problems,
e.g.: "He needs a lot of practice"; "give him
additional material to read"; "encourage him to work",
etC.

Much of the practice prescribed is carried out on an
individual basis with the teacher. Here agin, the
remediation is expressed in very general terms: "He
will work with an aid on short paragraph stories";
"spend time with him on an individual basis" (6);

”do more individualized work with the child" (9);

"have him read on a one-to-one basis" (8).

144

The data obtained from the list of all remedial

strategies mentioned by the teachers (whether applicable

to the two cases they described or to only one of them) not

only support the notion of a "general remediation" phenomenon,

but reveals three additional points:

1.

At least seven (70 percent) of the ten teachers stressed
psychological characteristics of the child: "I didn't
hassle her" (3); "encourage her to be brave" (4);

"make him feel comfortable about himself" (6), etc.
Thus, it appears that these teachers tend to emphasize
personal factors which might affect the reading act,
rather than offering specific remedial strategies to
overcome the problems.

At least five (50 percent) of the teachers said that
they send students like those described to reading
specialists for remediation (or that they at least
confer with specialists). Again, the teachers do this
in lieu of offering some specific remedial techniques.
At least four (40 percent) of the teachers, when asked
about remediation for a specific area of reading
difficulty, such as word recognition or word analysis,
said that they do not know what to do ["I don't know
where I'm going from here" (1); "I would have known at
least what not to do" (2); "the problem was never really

remediated" (3); "I don't know" (5)].

145

Summary: The remediation plans offered by the
teachers for the cases of reading difficulty reveal three
general trends:

1. The teachers seem to offer remediation at a very global
level. Emphasis is placed, specifically, on practice,
drill, and individualized help from others.

2. The teachers often refer students to reading specialists.

3. They try to help the students overcome some perceived
personal problems.

If these are the major remediation plans offered
by teachers for cases of reading difficulty, it must be
asked how successful are they in remeidating children with
reading problems in their classrooms? Following are the
teachers' responses, in their own words, to the question:
How do you know when and if you have remediated a child's
reading problem? (Question No. 12 in the interview)

Teacher 1: "I feel that if a child is able to read orally

a page, or silently, and be able to recognize or

to sound out words and at least answer little

questions...the child is on his way to really
learning to read."

"I never felt that I've (remediated). I don't

know. I never reached the point where I felt

a child has been remediated. I can see improvement

perhaps, or growth, but I don't know of anyone
that's ever reached that point."

Teacher 2

"When I see a change of attitude, that they are
liking reading, beginning to participate in class
work, raising their hand, answering questions
better, when things like initial sounds (that
were difficult for them at the beginning) become
easy, and the testing in general."

Teacher 3

Teacher 4:

Teacher 5

Teacher 6:

Teacher 7:

Teacher 8:

Teacher 9:

Teacher 10:

146

"I start to hear it: a) when a kid starts to make
sense; b) when (a child) starts to make pretty
close graphic or graph0phonemics correspondences:
c) when the kid starts to pull all things
together."

"a) If they can do exercise or practice after a
lesson; b) I guess with a number of kids I'm
aware of some specific problems that they had
and so I try to notice if they still have them."

"For some you can never remediate, that's just
their ability. For the average child that has
difficulty just a lot of repetition and drill."

"I have work sheets that I give out in the
afternoon and usually they correspond with
something the child has read about. If they
can do the worksheets then I assume the
problem has been remediated."

"a) When it comes time to change books or
through a test that I may give at the end of
the book and see how well they do; b) I try to
see if a child can orally read a page without
any or few mistakes. I (also) check back in
his workbook and see if he made a lot of
mistakes."

"I don't know if I've ever run into that.

I guess any kid that I've had that's really
been a low reader has still been a low reader
at the end of the year."

"In terms of specific skills: the reading

lab has individual tests for different objectives
at different grade levels. So after the child
has special instruction in a reading skill they
get this test and if they get them all right then
they figure the child knows the skill. If he
misses -- they go back and reinstruct and have
more work sheets."

Several characteristics of these responses stand out:

1. They range from general statements (e.g., "when the kid

starts to pull all things together" (4) to more specific

ones (e.g., "if they are able to recognize or sound out

words"

(1) .

147

2. They range from statements about non-measured
attitudes (e.g., "if I see a change of attitude, that
they are liking reading" (3) to statements about
observed and measured performance (e.g., "if they can
do the worksheets" (7).
3. They range from statements which express teacher's
confidence in remediations (e.g., "I start to hear
it" (4), to those which suggest a lack of confidence
in remediation ("I never felt that I've remediated"
(2). Three of the ten teachers expressed a similar
lack of confidence in their ability to remediate reading
problems.
Teachers' responses to specific interview ggestions:
QuestiOn No. 8: What overall methods do teachers use
in the classroom to find out chiIdren's reading performance?
Teachers' responses to this question indicated that,
in general, there are two general approaches to assessing
a child's reading performance: (1) to consider his or
her grouping from the previous year, with some additional
testing (3,9,10); and (2) to listen to the child read a
passage (1,4,5,6,7,8,10) and to pay special attention to
comprehension and vocabulary. Almost never are specific
tests or detailed diagnostic techniques mentioned.

Question No. 9: How do teachers teach reading
in their classrooms?

 

VOcabulary and comprehension, which were emphasized

in answers to the previous question, were also emphasized in

148

the teachers' responses to this question. These two
aspects of reading are apparently central to the reading
programs of all 10 teachers.

Specific strategies for teaching reading were rarely
mentioned; most teachers discussed what and how they teach
only in global terms (exceptions were teachers 6 and 9 who
outlined, more systematically, the different activities
they use in a typical reading lesson). This lack of
specificity might be attributed to two factors: (1) failure
of the interview to probe the more specific steps involved in
reading instruction, and (2) absence of a model which enables
teachers to more thoroughly formulate for themselves (and,
thus, to describe to others) the act of reading and its
various components.

Question No. 11: Do teachers individualize
readingrinstrucEIOn?

 

 

Most of the teachers said they do not individualize
instruction. Teacher No. 3 said she would individualize
only for readers in the lowest group, but not for those
in the top and middle groups. Teacher No. 8 emphasized
that, although she does not individualize instruction, she
does individualize the workbooks. Teacher No. 10 was the
only one to say that she individualizes reading instruction
"almost completely".

These responses seem to contradict some of the

findings discussed earlier, based on data obtained during the

149

initial part of the interview, when the teachers were

asked to talk about their remedial practices. As stated

(p. 143), at least three of the teachers (6,8,9) said

they work with students on an individual basis. There is

a two-part explanation of the teachers' apparent

inconsistency:

1. Usually, the teachers did not think of themselves as
individualizing reading instruction. However, during
the interviews, when they were encouraged to focus on
an individual child, they noticed that they sometimes
do, in fact, work with him/her on an individual basis.

2. Although the teachers do not individualize reading
instruction, they do individualize remedial reading.
In other words, there is a distinction between
instruction and remediation. Instruction is usually
provided to the class as a whole (or to groups).
However, those in need of remediation work with the
teacher on an individual basis.

Question No. 13: How do teachers know when
to refer a child to a reading specialiEt?

 

 

Teachers' responses to this question can be divided
into two:

Three teachers indicated that they do not refer the
poor reader to a reading specialist. They said they feel
they are qualified to help the student overcome his/her

problems (4,6,11).

150

Of the seven teachers who said they do refer
students to specialists, six explained that they make the
move when they feel they have done everything they can with
the students, but the reading problems persist. The
seventh teacher in this group, no. 9, said she does not
know much about reading remediation ("even though I teach
reading every day"), so she refers children after only a
week or so. (The six others indicated that they spend more
time collecting informal data about the reader before
deciding that she/he must be referred. Teacher no. 3 stood
out in this regard. She said she waits until January to
see whether a child has learned anything before sending him/
her to the specialist.)

Question No. 14: What do teachers think the
reading specialist should do?

 

Six of the ten teachers said that the specialist
should work with teachers -- as a consultant -- on the
diagnosis and remediation of reading problems. One teacher
(no. 3) even claimed that, in many cases, this consultation
should take the place of work between student and specialist.
She explained that there are too many students with reading
difficulties for the reading specialist possibility to
accommodate.

At least four teachers said that the reading
specialist should conduct diagnoses. Some explained that

they, as teachers, do not have the expertise to do

151

sufficiently thorough diagnoses. One teacher (no. 4)
called upon the reading specialist to do a "decent and
accurate diagnoses", instead of "employing standardized
testing procedures with all children". Furthermore, this
teacher said the reading specialist should "address
instruction to remediate the deficit", instead of
recommending some standardized remediation (e.g., "every-
body should do phonics").

Five teachers (5,7,8,9,10) said that the reading
specialist should also work on remediation, particularly
with students in great need of help. An important comment
made by teacher no. 9 was that the specialist should
also work with gifted children.

Overall, the teachers identified several duties and
responsibilitieis of the reading specialist:

1. They agreed that the specialist is an important
resource person in the school. (Some teachers said
they wished there could be even more specialists in
the school.)

2. They said the specialist should divide his/her time
between students and teachers; the teachers
explained that students are not the only ones who need
help in reading related activities.

3. They suggested that, because there are more deficient
readers than the specialist can accommodate, he/she
should work with only those students who are in great
need of help. This suggestion is consistent with two

findings reported earlier:

152

a. That even when instruction is not indi-
vidualized in the classroom, remediation is (p. 149).
b. That most teachers refer students to a specialist only
when they feel they have done everything possible for
the students, but have not succeeded in remediating the

problems (p. 150).

Question No. 15: What do teachers feel are
their responsibilities as reading teachers?

 

 

Because teacher's responses to this question were
diverse and highly interesting, they are provided verbatim:

Teacher 1: "Getting the children who are below grade
level as near to grade level as possible. Those
who are at grade level -- to enrich what they
already have."

Teacher 2: "Provide the child with appreciation for reading.
Develop in children the sense that reading tasks
vary from pleasurable reading to reading for
specific purposes. Encourage students to be
involved with the reading act. Provide extra
assistance to students who are showing some prob-
lems."

Teacher 3: "Get children to like reading. Get them to
become used to the system ("there is a system in
reading, once they become comfortable with it,
they have learned to like to read")."

Teacher 4: "Work on inferential understanding. Build the
children's vocabulary. Get the children to
understand that they are reading for two reasons:
to get information and to enjoy it."

Teacher 5: "To be aware of specific problems. To individu-
alize instruction. To get through the books by
June."

Teacher 6: "To do the best in teaching kids how to read."

153

Teacher 7: "To teach dhilren to read at a fundamental

level.

Teacher 8: "That at the end of the year a majority of the

children can go into a third year with no
problem."

Teacher 9: "Give the children positive attitude toward

reading. Introducing them to different kinds of
reading materials. Help them feel successful in
whatever they are in their reading (even if they
are below grade level). Make sure they have the
skills taught by the materials or program we
use."

Teacher 10:"To determine where the kids are in terms of

reading skills and level when they come to me.
Make sure that they are moving from there at
a pace that seems apprOpriate for each of them."

Several characteristics of the teachers' answers

are apparant:

a)

b)

C)

They range from very general statements (e.g. "to do

the best in teaching kids how to read" -- teacher (6)

to more specific ones (e.g., "build the children's
vocabulary" -- teacher (4).

They sometimes deal with students in general (e.g., "to
teach children to read at a fundamental level" -- teacher
(7) and sometimes with individual children (e.g., "to
individualize instruction" -- teacher 5).

They range from specific statements about reading
activities at the cognitive level (e.g., "work on
inferential understanding" (4) to more general statements
about the affective domain (e.g., "get children to like

reading" (3).

154

d) Several make no reference to the final product, or
results (e.g., "encourage children to be involved
with the reading act" (2), while others do take such
things into consideration (e.g., "that at the end of
the year a majority of children can go into a third
year" (8).

It is interesting to note, in this context, that
one teacher (no. 5), who had as one objective "to get
through the books by June" later said that the interview
made her aware that this objective is notso important. It
is more important, she said, to help children overcome their
reading difficulties. Accordingly, she added, "I begin to
feel that diagnosis is much more important than what I have
previously thought."

Question No. 16: Which of the following do_you feel

best characterizes your classroom situatiOn
with regard to reading instruction?

 

 

 

Four teachers (l,3,4,6) said that they have enough
time to diagnose all the students in their classroom (A).
Three (5,8,10) said they have enough time to diagnose
individual children who have reading difficulties (B).
Two teachers (2,9) said that, because of time pressure, it
is impossible for them to conduct individual diagnoses (D).
One teacher (7) was not sure of how to characterize her

diagnostic patterns (E).

155

These findings suggest that teachers do tend to view
reading diagnosis as a prime responsibility and that they do
take the time to diagnose children with reading difficulties.
It appears that only those who feel absolutely unqualified
to conduct diagnoses refer students to reading specialists
(e.g., teachers 2 and 9, who also mentioned earlier that
they do not know how to diagnose) (pp. 145-146).

These findings raise two important questions:

(1) How good is the diagnosis conducted by teachers who do
not refer children to reading specialists? and (2) When
teachers take the time to attempt diagnoses with slow
readers, how much time is left for them to work on reading

instruction with the other students in the classroom?

Summary

Results of this study were divided into three
parts: findings derived from product measures, findings
derived from process measures, and those derived from
classroom interviews.

The most significant finding suggested by the product
measures was that there is little agreement among teachers
on their final stated diagnostic judgments for given cases
of reading difficulties. This finding was supported by an
analysis of the number and type of diagnostic judgments
each teacher made, as well as by measures of the relationships
between the diagnostic judgments of each of the teachers

(diagnostic agreement score).

156

The major conclusion suggested by the process
measures was that teachers apparently lack comprehensive or
systematic approaches to gathering and evaluating
information about cases of reading difficulties. This
conclusion is supported by (1) differences among the
teachers on length of interactions with the cases, number
of cues collected and final number of diagnostic judg-
ments made, and (2) the type of verbatim comments they
made about the cues and the cases.

The major finding derived from the classroom
interviews was that teachers consider similar global
diagnostic categories in both laboratory situations and
the natural classroom environemnt. It was also found that
diagnostic and remedial categories mentioned by teachers in
the classroom tend to be very general. This may indicate
an absence of a comprehensive functional model of reading
and diagnosis. It may also raise questions about the
adequacy of the training teachers receive in diagnosis and

remediation skills.

CHAPTER V

SUMMARY, CONCLUSIONS AND RECOMMENDATIONS

Introduction

 

This study was an exploratory one designed to
investigate the clinical problem solving behavior of
classroom teachers as they diagnosed students' reading
difficulties in experimental and classroom situations.

The study was part of an extensive research program undertaken

by the Clinical Studies component of the Institute for

Research on Teaching at Michigan State University.

The literature on reading diagnosis indicates that
classroom diagnosis is viewed as an essential part of all
reading instruction. More specifically, it suggests the
following:

1. There is an agreement among educators that reading
difficulties hinder students' intellectual and emotional
development, and that early and continued classroom
diagnosis is needed to prevent and remediate the
reading problems.

2. There are varied meanings associated with the term

"reading diagnosis." Most definitions suggest that there

157

158

are different levels of diagnosis: classroom, school, or
clinical. However, missing from all definitions are
descriptions of the cognitive processes teachers employ
as they gather information and attempt to reach
diagnostic decisions.

3. Although most researchers believe classroom teachers are
capable of diagnosing students with reading difficulties,
there is not enough empirical data to confirm or
disconfirm this belief.

In summary, a review of the literature failed to
provide a description of the diagnostic processes classroom
teachers use. Moreover, the review revealed a lack of
empirical evidence about the competence of classroom

teachers in diagnosing reading problems.

Procedures

 

Ten teachers were involved in the study, five from
the Chicago area and five from the Lansing area, all had
taken at least two courses in reading instruction at a
university level.

All 10 teachers went through the same procedures.
First, they interacted with two simulated cases of reading
difficulties in a laboratory situation. These simulated
cases, developed by Dr. Rebecca Barr of the University of
Chicago, represented the behaviors of students with reading

deficiencies. The cases were based on materials that would

159

be available to a classroom teacher and made use of

the students' natural language patterns. In a second

phase of the study, the teachers were observed in their own
classrooms, and interviewed about their instructional,

diagnostic, and remedial practices in reading.

Data Analysis

 

Data from laboratory interactions consisted of
verbatim comments made by the teachers as they gathered
and evaluated information about the two simulated cases,
and of their final stated diagnostic judgments. These
data were analyzed by means of product and process
measures. Product measures consisted of (l) diagnostic
agreement scores, (2) diagnostic ommonality scores, (3)
cue agreement scores, and (4) cue commonality scores.
Process measures included (1) length of interaction, (2)
number of final diagnoses, (3) number of cues collected, and
(4) relationships between cues and the teachers' verbatim
comments.

Classroom interviews were analyzed in terms of (l)
diagnostic categories mentioned in the classroom, (2)
remedial techniques offered, and (3) teachers' responses
to various questions in the interview. Comparisons were
also made between teachers' classroom diagnoses and their

laboratory diagnoses.

160
Results

The major finding suggested by the product measures
was that the teachers differed from one another on their
final stated diagnoses for the simulated cases of reading
difficulties. The process measures yielded two major
findings: (1) that the teachers apparently lacked comp-
rehensive and systematic strategies for gathering and
evaluating information about the cases in order to reach
diagnostic judgments, and (2) that the teachers differed
markedly from each other on length of interaction with
a case, number of cues collected, number of final stated
diagnoses,and.number of comments made about case.

Analysis of classroom interviews revealed that the
teachers mentioned similar global diagnostic categories
in both the laboratory situations and the natural
classroom environments. It was found that these categories

tended to be very general.

Conclusions

 

Despite the small number of teachers participating,
the study provides convincing evidence for the following
conclusions:

1. Contrary to what the professional literature on
reading diagnosis implies, classroom teachers do not
seem qualified to diagnose and remediate cases of

reading difficulties. The results of this study suggest

161

that teachers are ill-equipped to carry out these
functions. This apparent deficiency may be the
result of two interrelated causes: (1) insufficient
training of teachers in reading diagnosis and
remediation, and (2) absence of a comprehensive
model of reading and diagnosis.

Several examples support this conclusion. As one
teacher (no. 7) interacted with a simulated case, she
said: "The problem is that (the child) can't have phonetic
analysis of words". Later, when this teacher stated her
final diagnoses, she focused entirely on emotional problems
(she also cited emotional problems during classroom
interviews). It seems that this teacher Operates from a
psychological model, which is only remotely related to the
reading act, if at all.

Another teacher (no. 8), after looking at only a few
items of information about a case, said: "From her word
meaning knowledge I assumed very fast that comprehension
was one of her better areas". In her subsequent work on
this case, this teacher looked at no items of information
about comprehension (whereas seven other teachers had
something to say about the comprehension level of this
case).

A third teacher (no. 9), interacting with simulated
case (Case 7), made two final diagnoses that contradicted
each other: (1) "the child reads hesitantly", and (2) "the

child reads very well".

162

These examples suggest that these teachers do not
operate from comprehensive models of reading and diagnosis.

The notion that teachers are ill-dquipped to conduct
diagnoses is also consistent with results from studies of
reading specialists and learning disabilities personnel
conducted by the Clinical Studies group. Tables 19 and 20
show the diagnostic categories most frequently mentioned by
eight senior reading clinicians who interacted with
simulated cases in a 1977 observational study (Vinsonhaler,
1979), and those categories most frequently mentioned by
the teachers in the present study.

A comparison of Tables 19 and 20 demonstrates that,
in general, results for the proportional agreement in the
two studies are quite similar. In both instances, most
of the diagnostic categories listed were mentioned for a
given case by only about 30 percent of the clinicians.
There were, however, more diagnostic categories cited by
at least 50 percent of the clinicians in the 1977 study
than in the present study, possibly because the clinicians
had more similar training backgrounds than did the 10
teachers. This explanation is consistent with the
Agreement Corollary described earlier (pp. 57-62).

The consistency of results between the two studies
is further illustrated by a comparison of the diagnostic
agreement and commonality scores for each. Table 21

summarizes these comparisons:

163

TABLE l9.--Most frequently mentioned diagnostic categories

for 1977 observational study

(8 clinicians,

4 cases)*

 

Diagnostic categories

 

At least average
reading potential

Adequate verbal skills

Poor oral reading

Problems with vowels

Sight words: Low

Phonics: Weak

Auditory acuity: Problem

Consonant blends: Not
a problem

Good use of context

Writing: Problem

Spelling: Problem

Normal interest and
behavior

Attitude toward reading:
poor

No problem isolated
letter sound skills

Speech: Problem

Problem with Syllables

Handwriting: Problem

Problem with visual
memory

Health problems in
school

Poor word analysis skills

Auditory discrimination:
problem

eases

.67
.33
.50
.50
.33
.33

.33
.33
.33

.50

.50

Case B

.33
.50
.67
.33

.50

.33
.33

.50
.50
.50

Case C

.50
.50
.33
.33
.83

.33
.33

.33

.50

Sea-“£2

.67

.67
.67

.33
.33

.50
.50

.50

 

*--denotes 0.00

164

TABLE 20.--Most frequently mentioned diagnostic categories

for Cases 7 and 8 and 10 teachers*

 

SELECTED DIAGNOSTIC CATEGORIES Case 7 Case 8
(54 total DX) (62 total DX)

 

Endings: Ignores .50 --
Sight words: Weak -- .50
Sight vocabulary: Good .40 --
Vocabulary concepts: Adequate .30 .30
Comprehension: Poor .30 --
Oral reading: Skips words -- .30
Oral reading: Does not pay attention

to punctuation -- 30

Phonic skills: Weak --
Word attack skills: Lacks --
Print translation: Problem area --
Oral reading comprehension: Good --
Understanding: Good when listening --
Literal comprehension: Adequate --
Beginning sounds: Knows --
Initial consonants: Strong --
Syllabication skills: Poor --
Unfamiliar words: No attempt to sound --
Context to help with the unknown word:

I O O O O O C O I O O
N
0

Does not use -- .20
Language: No language problem -- .20
Comprehension: Adequate .20 --
Vocabulary: Weak .20 --
Punctuation: Lacks .20 --
Reversals: Problem area .20 --
Mispronunciation errors: Did not

correct .20 --
Vowels: Need work .20 --
Word analysis: Weak .20 --

 

*--denotes 0.00

165

TABLE 21.--Diagnostic and commonality scores for the
classroom teachers study

and the 1977 study

 

Classroom teachers study 1977 Study

 

 

 

 

10 Classroom teachers 8 Reading Clinicians
STATISTICS Case—7 Case 8 (24 data points)
score SD score SD score SD
Diagnostic
agreement -.04 .13 -.03 .11 -.07 .17
Diagnostic
commonality .45 .29 .45 .23 .55 .18

 

 

 

As Table 21 shows, diagnostic agreement and
commonality scores were markedly similar for the classroom
teachers and the reading clinicians. This further
suggests that the findings of the present study are
generalizable.

2. Overall, teachers seem to lack the information processing
strategies that are necessary for gathering and
evaluating information about students' reading
difficulties. Their diagnoses in both the laboratory
situations and the natural classroom environment tend
to be very global, non-specific and incomplete. They

are confined to a set of general observations that do

166

not necessarily form a comprehensive picture of
the child's reading performance. The teachers are
observers of students' behaviors; at best, they
can say what the children cannot do, rather than
offer complete and specific diagnoses.

That this is so should come as no surprise, for

several reasons: For one thing, teacher education programs

rarely provide teachers with thorough preparation in

decision-making and diagnostic skills. Moreover, the

models of reading and learning teachers are given in training

programs may be inadequate and incomprehensive. Finally,

most programs emphasize method courses and the acquaintance

of prospective teachers with a variety of subject areas,

rather than more specialization.

3.

Investigations conducted in laboratory settings,
carefully planned and monitored, seem to be a valid

and apprOpriate means of studying the clinical

problem solving behavior of classroom teachers. The
fact that the performance of the individual teachers

in this study were consistent across two simulated
cases suggests that the cases elicited at least some of
the teachers' habitual problem—solving behavior. That
the general diagnostic categories mentioned by teachers
in the laboratory settings were also mentioned during
classroom interviews suggests that the teachers

operated from similar models of reading in both

167

situations. Furthermore, the teachers' own
comments about the laboratory interactions support
the claim that the laboratory conditions did,
indeed, resemble the natural classroom situation

with regard to the diagnosis of individual readers.

Recommendations

 

There is a clear need for the classroom teacher to
be a reading diagnostician. However, training in the
diagnosis of reading difficulties has been reserved mainly
for graduate level courses; most undergraduate teacher
education programs do not provide sufficient instruction
in this area. The adequacy of this kind of teacher
education curriculum may therefore be questioned.

Teachers need to be trained more thoroughly in diagnostic
and remedial skills in reading. Such training should be
based on a model of reading and diagnosis, and focus on
the diagnostic strategies involved in gathering and
evaluating data about students with reading difficulties.

Training of teachers in diagnosis and remedial
skills can be conducted in laboratory situations, using
simulated cases of reading difficulties. Teachers should
be given much practice interacting with simulated cases,
where it is possible for them to receive immediate feedback

on their performance.

168

It is also necessary that additional studies be
carried out to investigate further the clinical
problem-solving behavior of classroom teachers in reading
diagnosis. Instruments should be develOped to relate,
more precisely, teachers' diagnostic performances in
laboratory situations to their performances in the
natural classroom environment.

Research should also be conducted to investigate
the impact of laboratory training with simulated cases
on teachers' performances in the natural classroom

environment.

APPENDICES

APPENDIX A

CASE INVENTORY

APPENDIX A
CASE INVENTORY

.xmcu coHcmnmumEoo mchcou Hmuo or» CH hHuomuuoo umzmcm
no: OHO acmosum mnu noHna mcoHumosU mmocu acmcsum mcu mxmn nonomou
on» coca can .mEHu ccoomn on» u0u command on» psoHc mommy ucmozum an» I mmomm onmzmmmmmzoo quadmm Admo
.xmmu concmsmumEoo mcHommu
Hmuo on» cH aHuomunoo umsmcm uoc ch ucmpsum on» :oHrz mcoHumosq mmocu unmosum
on» wxmm umsomou or» cocu .ommmmmm m mchcmu morons» on» o» mcmumHH unocsum on» I onmzmmmmmzou oszmemHH

 

“mmomm ZOszmmmmmSOU

.uH usond mcoHummsq uozmcm Ou coxmm mH one OOOHIuso command mchomu m mommu unoccum on» I onmzmmmmmzoo wzHo<mm Haze
.uH usonc mcoHuwmsq Hmuo umzmcm cu cmxwm
nH coca can NHucmHHm mmmmmmm m mecca ucmcsum may I mmmzwz< can monammDo ammo :uHB onmzmmmmmzou ozHo<mm azmHHm
.uH usonm mcoHummoq cmuuHuz um3mcm ou cmxmm

OH cmnu can >Hucmdfim wmmmwmm m mvmmu ucmvsum On» I mmm3m2¢ flaw mZOHBmmDO ZMFFHm3 nufl3 ZOHmzmmMM&ZOU OZHD¢mm BZMQHm

 

.mcoHummsq OHmwommm admummummIco: no smcummucm OHMHommm m cquH3 mcoHummsq uo umHH ”onmzmmmmmzou

.uH mocsocoum o» ucmpsum mcu mxmm
can msoum umuumH uo umuumH onch m mucHum morocco mnu I AmwnaHmcm OHumcocmv onecHUCmmm Homzwm ozoom
.uH mocsocoum ou ucmcsuw
mnu mxmm can» can mOHOMHHmm OucH puo3 coucHud m mmcH>Hc umncmou on» I AmmumoHanHam accommu. quxzumU «aromas
.uH mocsocoum
0» coca can mmHanHam Oucw ouoz OOOCHum c mcH>Hc Op coxmm mH ucocsuw on» I AmmumOHOMHHNm ucmpsum. UZHXZDmu Bzmoaem
.uH mocoocoua ou campsum or» mxmm can Ono: m mucHuQ umnommu on» I onedqomH

 

AZOHfiflloHBZMQHV ZOHBHzooumm Qm03

commune mchmmu one CH musuoHd on» NO coHumHuOmmn
commune OCHpmmu on» mcHhccmEooom musuOHm

command mcu mcHocmn MHuoHu co mucoEEoo
coHumHuomcnuu OCHpmmu cho

mucoEEOO m.noHOmcsoo Hoonom

coHucEMONCH umma

>uouan Hoozom

Emumcrm OCHOmmm

Hoocom CH mocmEuOunom m.MHuoHu

ccsoumxomn mEonIHmusuHsU

DZDOmUxU<m

.ouos can mcmec ou ucmcsum on» mxmc con» one mocmucom m CH puo3 c moms nonommu may I muzmezmm ome<mmzmo «aromas

.oocwuemm m :« cuoa m was on taxes mH unmesum we» I mozmezmm ameammzmo azmoaem
.coHuoHOmH :H EH: 0» poucmmoum cuoz m OCHump ou poxmm OH ucmpsum man I onedqomH

 

HmBQMUZOU Hm<HDmQUO>V UZHdez QmOZ

 

meBzm>zH mm<U

1159

APPENDIX B

DIRECTIONS FOR OBSERVATIONAL SESSION

APPENDIX B

DIRECTIONS FOR OBSERVATIONAL SESSION

I. Introduction

This research is being carried out by the Institute
for Research on Teaching (IRT) at Michigan State University.
The IRT needs to observe a number of classroom teachers in
order to develop some ways to train teachers to diagnose
children's reading problems in the classroom. By diagnosis
we mean: to determine the most probable strengths and
weaknesses of the child's reading performance. You have been
chosen as one of those who will be observed during our study.
Because this work will take several hours, the IRT will pay

you as a consultant for the College of Education.

II. Orientation

Let me explain the observational session. First, I
should emphasize that all personal information regarding this
session will be kept confidential. Your name will not be

part of IRT permanent records. Instead, a number will be

170

171

used. We are required by law to protect your privacy by
keeping confidential your name, social security number, etc.
Second, I should emphasize that we are not evaluating you in
any way. we are merely interested in understanding how you
go about determining the reading problems of a given child.
Finally, let's consider what we shall actually do. The
observation will be divided into two parts. In part I, you
will interact with materials to analyze a simulated case of

a child having reading difficulty. The case includes audio
recordings of the child reading a passage aloud, a written
transcription of his oral reading, some background information
about the child and different tasks the child performed (e.g.,
his answers to comprehension questions). In just a minute I
will explain to you how we developed this case. We would
like you to interact with this case in the same manner you
would use in interacting with a real child in your classroom.
Please think aloud and verbalize your analysis of this case
as you diagnose it. Stop whenever this interferes with your
work.

During part I, an audio tape record will be made as
you examine the case. The equipment here is required to
prepare the tape. The tape recording will remain completely
confidential. In part II, we shall jointly review the case
while you attempt to recall what you were thinking about as
you attempted the diagnosis. An observer with a reading
background will join us to provide you with a "sounding board"

to help you clarify your thinking and get a record down on

172

paper. Do you have any questions? (pause a few seconds;
answer questions by repeating instructions or postpone
discussion). (Experimenter collects subject release and

consultation forms).

III. Case Inventory and Practice Case

We will now begin the session. Information concerning
a case has been prepared for you to look at. Your task is to
request this information about a case and to use this in-
formation to determine the most likely diagnosis about the
child's reading performance and to suggest a general program
of remediation. The way we developed this simulated case is
as follows: we surveyed the children in a classroom and we
picked up a child having reading problems in the classroom.
We asked the child to read a printed passage, and then we
collected some basic information on the child that seems to
be useful to a classroom teacher in order to figure out what's
wrong with the child's reading. In just a minute I will give
you a copy of the reading passage. (Experimenter shows to
the teacher the reading passage). The different types of
information available are: Background (experimenter points
to Background in the Case Inventory); Comprehension
(Experimenter points to Comprehension); Word Meaning
(Experimenter points to Word Meaning); and Word Recognition

(Experimenter points to Word Recognition). As a teacher you

173

probably deal with these concepts on a daily basis in your
classroom. However, the way in which they are used here may
be slightly different from the way you typically use them.
So, I would like to review with you in a systematic manner
these concepts. This Case Inventory is only for training
purposes. This is not for the case that you will be asked
to diagnose (pause for a few seconds).

Let me explain to you in greater detail what you can
find under each one of these types of information. If you
turn the page to Background, (Experimenter waits for the
teacher to turn the page to Background) you can find several
types of information as listed on this page (Experimenter
points to the list; pause for a few seconds). Do you have
any questions? (Pause a few seconds). OK, if you turn now
to Comprehension (Experimenter waits for the teacher to turn
the page to Comprehension) you can see that you have two
major types of information: Comprehension (Experimenter
points to Comprehension) and Comprehension Probe (Experimenter
points to Comprehension Probe). On pages 2 through 4 you can
find definitions and examples of these types of Comprehension
tasks. You may take a minute to examine them (pause for a
minute). Do you have any questions?

If you turn now to Word Meaning on page 5
(Experimenter waits for the teacher to turn the page to Word
Meaning) you can see that the information is presented here

in three verbal forms: in Isolation (Experimenter points to

174

Isolation), in Student Generated Sentence (Experimenter points
to Student Generated Sentence) and in Teacher Generated
Sentence (Experimenter points to Teacher Generated Sentence).
You may take a minute to examine these types of information
on page 6 (pause for a minute). Do you have any questions?
(answer questions as necessary). If you turn now to Word
Recognition on page 7 (Experimenter waits for the teacher to
turn the page to Word Recognition) you can see that the
information here is presented in several forms. Please take
a minute to examine these forms on pages 8 through 12. Let
me know if you have any questions. (pause a minute).

The last page in this Inventory (Experimenter waits
for the teacher to turn to the last page) is a summary of all
the types of information that we have just reviewed in this
booklet. This summary is identical to the booklet. To save
you time you may use this summary. You may take a minute to
re-examine these types of information. (pause a minute).

Let's discuss now how you can request to see a
specific type of information. When you diagnose the case, you
may request information by referring to the Inventory.

This is the reading passage for the practice case
(Experimenter hands out to the teacher the reading passage for
Practice Case). All words underlined with (wr) (Experimenter
points to "laid") can be requested for Word Recognition (pause
a few seconds). All words underlined with (wm) (Experimenter

points to "baker") can be requested for Word Meaning (pause a

175

few seconds).

Now about comprehension questions about the reading
passage. We could not anticipate all questions that you might
ask and solicit responses from the student. So we prepared
lists of possible questions according to paragraphs in the
reading passage. When you want to ask questions relating to
a specific paragraph, tell me which paragraph it is and
whether you would like to ask questions related to listening
comprehension, oral reading, reading probe or silent reading.
You will then receive a list of questions to examine. For
example, if you ask to see questions for listening compre-
hension from paragraph 1 you will receive this list
(Experimenter hands out C-L-paragraph 1). Now, if you want
to see, for example, the student's response to question number
3, tell me it and I will give this information to you (Experi-
menter hands out C-L-Pl-3 ; pause a few seconds). You can
ask to see a list of questions which are non-paragraph specific,
but be sure to indicate whether you want to see oral, silent
or listening questions. Do you have any questions? (pause a
few seconds).

To help you become familiar with this procedure, we
will let you work for a few minutes on a practice case. This
case is not one which you will be asked to diagnose. We just
want you to have an opportunity to examine forms of information

listed on the Inventory without attempting to diagnose the

case. You have the Cue Inventory for your practice case.

176

You may now request a couple of items of information
from the Inventory. I will assist you. (limit of 2 items

from each one of the four major types of information).

IV. Observational Session

(Take back all materials for practice case). Here is
the Case Inventory for the case you will diagnose. Take a
few minutes to study this Inventory before we proceed (hand
out Inventory and pause for 30 seconds). You can request as
many items as you wish, but you must request them one at a
time. There is no right or wrong amount of information to
request for your diagnosis. Use the same procedure you
would use with a child in the classroom. When you request an
item of information, I will give it to you. You may keep all
items requested throughout the session. Do not feel you
must request an item of information because it is present
in the Inventory. Do not feel you must request items in the
order listed in the Inventory. Do you have any questions?
To review, (hand subject the summary of instructions)
the instructions are as follows:
1. A new girl has just joined your classroom and you
want to assess her reading performance.
2. Select information you want from the Inventory
and ask me to give it to you.

3. When you are ready summarize your opinions about

177

the child and what you may do for her in the class-
room.

4. Verbalize your thoughts. If it interferes with
your work, stOp verbalizing.

Ok, let's begin. Our case concerns a _____year old

(girl) named .

 

You may take a maximum of one hour for your diagnosis
and 30 minutes to write down your diagnostic Opinions. Here

are some initial items of information.

Post observational session instructions

 

To finish up, we would like you to summarize your
judgments in a written form. Please briefly state your
diagnostic opinions and your suggestions for remediation
on these sheets. Be as specific as possible. In writing
your diagnosis and remediation assume that the report will be
used either by you to work with this child or by a reading
specialist to whom you may refer the child. Please write
as clearly as possible. You may take a maximum of thirty

minutes.

V. Debriefing Session

We now will begin part II of our session. The

purpose is to clarify for us the way you went about making

your decisions. To help you we will review the session with

178

(CO first name) who has observed you at work. We will hand
you the items of information in the order in which you
requested them. Then (CO first name) or I will ask you
questions or let you comment about your thinking. The purpose
of this procedure is to stimulate your recall of your thinking.
As we go along (CO) will record your comments on a data
record which will be used for later analysis.

Ok, let's proceed (Comment to experimenter): For
each time/cue entry on the observation record, use the
following procedure:

1. Why did you ask for this piece of information?

(If subject's response includes items 2 and 3
or 4, stop).

2. What did it tell you, if anything? (If subject's

response includes 3 or 4, stop).

3. Did you already have a hunch that this information

confirmed or ruled out?

NO - (GO TO 4)

YES - What was it?

Was it confirmed or ruled out?

Stop!

4. Did this information give you any new hunch?

NO - Stop!

YES - What was it?

(count as GENERATION StOp!)

Instructions on completion of debriefing for

diagnosis:

179

E. To complete our work, we would like to give you a
chance to re-write or modify your diagnostic
and therapeutic decisions -- just in case the
debriefing has changed any of your ideas. (hand
subject a new sheet of paper for additions or
changes).

CO: Underline each of the diagnostic categories

mentioned in the written diagnosis. For each of

these terms, CO asks the following questions:

1. Would you define as clearly as possible what you
mean by this term?

2. Can you think of any synonyms for this term?

3. Can you give me some behavioral examples of a

child with this problem?

APPENDIX C

CLASSROOM INTERVIEW

APPENDIX C
CLASSROOM INTERVIEW

Experimenter: (Shows teacher descriptions of four
children with reading difficulties, two of which are
identical to the two simulated cases diagnosed by the
teacher in the clinic situation):

Here are four descriptions of real children with reading
problems. As you read through each description, try to
see if you can think of a child in your classroom who
generally fits the description. Could you please think
of children in your classroom who correspond to each of
these descriptions? Teacher: (Identify children in the
classroom).

Exp: (Focuses on one identical child. When possible
select a child who corresponds to one of the two descriptions
which is identiCal to the simulated cases, ask the teacher
to reread the description):

Is there anything missing from this description that you
can find in your child?

Exp: Is there anything in this description that you

do not find in your child?

Exp: In general, what did you do to find out that this
child has these problems?

Exp: (For each one of the categories in the description,

ask the teacher):

180

181
A. Why do you think the child in your classroom corresponds
to this description? (or: has this problem?)
B. What sources of information have indicated to you that
she/he has this problem?
(1) Oral reading
(2) Standardized tests
(3) Reading related observation
(4) School records (specify)
(5) Practice activities
(6) Other (specify)
C. Why did you collect the above information (cue)?
(1) Always collect this information
(2) Because it is important in teaching the child
to read
(3) Not teacher initiated (explain)
D. Why do you think this information (cue) indicates
that the child has this problem?
(1) Reasoned in terms of a Model [pupils who can't
do this always have this (problem)]
(2) Reasoned in terms of experience with other
children
(3) Reasoned in terms of authority (general references
to authority (e.g., a book, a course)
(4) Other (specify)
E. What do you intend to do with (child's name) to
remedy his reading problems?

(Exp: Proceeds to the next category in the description

10.

182

and repeats questions A through E).

Do you have any of the following materials?

A. Tape recordings of the child reading a passage

B. Oral reading transcription prepared by the teacher

C Sample of text used in oral reading

D. Exercises done by the child and used by the teacher
in diagnosis

E. Exercises given to the child as part of the remediation

F. Other (specify)

(Exp: Starts again with question number two and selects

a child. If possible the second choice corresponds to

the second description which is identical to a simulated

case. Repeates question two through six).

Exp: Can you summarize for me the overall method that you

use with children in your classroom to find out their

reading performance?

Could you briefly summarize how you teach reading in your

classroom?

(Exp:

A. Asks specific questions in relation to what has been
observed during the reading session, or

B. Describes to the teacher what has been observed
during the reading session and asks if the teacher has
anything to add to it).

(Exp: To finish up, let me now ask you some general

questions):

183

11. Do you individualize reading instruction?
12. How do you know when and if you have remediated a child's
reading problems?
13. How do you know when to refer a child to a reading
specialist?
14. What do you think the reading specialist should be
doing?
15. Could you please tell me about what you feel are
your responsibilities as a reading teacher?
Exp: Could you please rank them in order of importance?
16. Which of the following do you feel represents your
classroom situation with regard to reading instruction:
A. I have the time to diagnose all the children in
my class.
B. I have enough time to diagnose individual children
who seem to be having trouble.
C. I prefer that children with reading difficulties be
seeing a specialist or consultant.
D. Time limitations make it impossible for me to do
individual diagnosis.

E. None of the above characterize my classroom.

APPENDIX D

DESCRIPTIONS OF CHILDREN WITH READING DIFFICULTY

APPENDIX D

DESCRIPTIONS OF CHILDREN WITH READING DIFFICULTY

MICHELLE

Word Recoggition

 

Although Michelle has the beginnings of a sight
word vocabulary, it is inadequate for her present grade
placement. On several occasions she confused visually
similar words such as "three" and "there" but after a more
careful examination of (these) words she was able to
self-correct her initial confusion and to make the fine
visual discrimination needed for accurate identification.
This skill area needs work and Michelle's complete attention

(concentration) needs to be brought to the task.

Contextual Reading

Michelle, in oral reading situations, appears to have
some facility with contextual prediction. She is able to
effectively use context as a mediating strategy to help her
get through a passage and to hold on to the sense of the
sentences. However, because Michelle has to stop to decode
so many words, her reading is reduced to the level of
word-by-word, very choppy reading with frequent disregard of

punctuation.

184

185

Word Analysis

 

In the area of phonetic analysis, Michelle demonstrates
a mastery of some sound units in isolation. With direct
assistance she can accurately blend together sound units.
However, she needs to gain more facility in blending. She
needs to move toward independent use of phonetic analysis in
the actual process of reading. Applying phonics to decode
a word is not automatic but she is moving toward this desir-

able level.

Reading and Listening Comprehension

 

Both oral and silent reading comprehension appear to
be low for present grade placement. The comprehension measures
available focused on factual recall of information, a lower
skill level in the hierarchy of comprehension skills. One
measure of comprehension potential suggests that she can
orally understand material at an apprOpriate level. This
suggests that instruction in comprehension coupled with
instruction in word recognition strategies will readily
enable Michelle to move ahead in reading comprehension.
Presently, her deficits in word recognition are holding back
her growth in reading comprehension. She cannot understand

what she cannot decode.
JIM

Synthesis of Findingg

Given extreme difficulties in print translation, Jim

186

extracts more stated information than might be expected. It
suggests that he uses his good meaning vocabulary and
conceptual strength to make sense of this passage on the basis
of unsystematic information derived from print. Thus, it

is concluded that his inferential comprehension skills are
well developed in spite of his failure to answer some of the
inferential comprehension questions correctly. His major area
of reading difficulty is in word identification and he may
also have a minor problem in the contextual processing of

sight words.

Word Recoqnition in Context

 

When orally reading a passage, Jim had difficulty
recognizing 19% of the words. Errors were evenly distributed
between high frequency sight words and other less common words.
Because of the difficulty of the passage, the results of the

analysis should be viewed with some degree of caution.

General Observations

 

On five words Jim dropped or added "s", which suggests
the influence of dialect. Few repetitions were made (7);
most of these were made (5) in order to correct a previous
miscue. One habituated response, "was" for "were", occured

three times and might also be a function of dialect.

Sight Words
While many sight words were known by Jim, the
pr0prtion with which he had difficulty was large (10%) and

tended to interfere with fluent reading. Many sight word

187

confusions did not seem to be the result of difficulty with
nearby uncommon words. All sight word substitutions (20/20)
were cued graphically, typically by initial consonants
(16/20). About two-thirds were also cued by prior sentence
context. But a large prOportion of the substituted sight
words (two-thirds) did not conform to the author's sentence
meaning. Because of the large number of sight word
substitutions and their frequent distortion of the author's
meaning, sight words are judged to represent a problem for

Jim.

Analyzed Words

 

Uncommon words were dealt with by Jim either by
omission (9) or through substitution (11). He seemed to
shift from not attempting uncommon words, to trying to
identify them, and then back again to omitting. Contrary to
his pattern with sight words where prior sentence context was
used only in conjunction with graphic cues, three of ten
uncommon words were cued only by prior sentence context.
Nevertheless his dominant strategy (50% of substitutions)
was to try to use cues from the prior context, from letter
sounds mainly at the beginning of the word, and word length.
The remaining two words were graphically cued. No noun
words were pronounced and there was no evidence that he was
attempting to break words into syllables except possibly for
the substitution of pic(k) for picnic. He appeared to be

using a fairly unrefined method of whole word substitution.

188

He seemed to try to compensate for his lack of skill in word
identification by attempting to make enough sense out of the
story to permit him to guess at words. (For example, plates
was finally identified and picnic was attempted after

several omissions).

Context-Comprehension

Many of his substitutions showed the influence of prior
sentence context. He used a correction strategy seven times,
usually to replace a substitution that didn't make sense.
Sometimes, however, the correction was based apparently on
graphic cues (e.g., there/these; maybe/men). Nevertheless,
there were instances when substitutions did not make sense and
Jim did not attempt to correct. Perhaps what he read did
not create sufficient meaning for him to have a basis for
correction. On the whole, he appears to be sensitive to
contextual meaning and, when possible, to use contextual

information as a cue in identifying words.
JOE

General Observations

 

Joe had difficulty with comprehension of verbal
material, whether reading it or listening to it aloud.
This conclusion is supported both by his difficulty in
answering questions about the story when he was unable to read
an answer from the text, and by his teacher's report that he

has difficulty understanding directions. His difficulty may

189

be compounded by poor memory, as evidenced by his incorrect
answers to questions (after the listening) which he'd

answered correctly when reading from the text. His teacher
also reports that Joe seems to have difficulty remembering

responsibilities at school.

Context Comprehension

 

The oral reading probe indicates that Joe can locate
answers to questions when the answers are stated literally
in the text. The process seems to break down at the point of
integrating this knowledge into a more comprehensive structure
into which he could channel his own experience. The lack of
such integrated understanding prevents him from drawing
inferences and impedes memory, since he would probably have
to hold in mind many virtually unrelated deatils instead of
interrelated components and events which influence each other
and have some meaning.

Joe seems to have the meaning vocabulary and
experience necessary to understand this story. While his
definitions are actually demonstrations and used in sentences,
they seem adequate to express his familiarity with the items.
He was able to infer an adequate meaning of the one word he
didn't know from the context of the teacher's sentence. This
evidence indicates that meaning vocabulary is not interfering

at this time with his comprehension.

190

Word Recognition in Context

 

Joe has some difficulty with print translation. Some
attention needs to be paid to his difficulty with endings
in context and to a system for attaching unknown words. The
fact that he could figure out words when helped to chunk
indicates that would be a profitable strategy to teach him --
coupled with visual analogy.

The difficulty with endings suggests that he is not
using contextual clues as well as he might. His low rate of
corrections also supports this. The "ed" ending seems to
present a special case of a misunderstanding about how this

ending affects the number of syllables in a word.

DIANE

General Observations

 

Diane appears to be a well-adjusted, sports-minded,
little girl. Her own verbal impression is clear, spontaneous,
and adequate for good conversation. Her associations with
peers, teachers and parents are positive and her health
records indicate no problems. Academically her I.Q. scores
indicate at least average intelligence, and she is experiencing
good success with school tasks with the exception of reading.
There is a bad attitude toward reading developing and possibly
a feeling of inability building. Here reading problem appears
to center around word recognition skills -- not comprehension.
Listening comprehension scores and one-to-one reading test

scores indicate good concentration/comprehension though her

191
attention span shorten when working alone on difficulty
tasks. Diane's reading problem is located primarily in
the decoding area. That is, if she could develop better sight
vocabulary and word attack skills she has the potential to be

an average or better reader.

Sight Vocabulary

 

Diane's problem areas seem to be concentrated as
follows: The most recurrent problem is the fact that when
reading a word she often looks only at the first letter (or
first few letters) in the word ignoring the middle and end.
From these few cues she guesses an identification -- many
times incorrectly. This makes her a dependent reader and so

she constantly asks for teacher's help.

Word Analysis

 

Upon an examination of her phonic spelling, it becomes
apparent that word chunks (-tion, -ing, -ack) are not present.
As these are much more regular than single letter sound

connections, they should be worked on.

Word Problems in Context

 

In looking at her wild guesses, it becomes apparent
that context cues are not being used consistently, and
considering Diane's language abilities and I.Q., this should
be worked on as well.

Lastly, due to her word analysis weaknesses, and low
sight vocabulary (which seems to be due to an insufficient

examination of the word), she struggles to decode. Because

192

of this she probably can't focus her attention on grammatical
markings because she is too busy focusing her attention on

figuring out words -- word-by-word.

APPENDIX E

FREQUENCIES OF DESCRIPTIONS OF READING DIFFICULTIES

SELECTED BY TEACHERS IN THE CLASSROOM

APPENDIX E

FREQUENCIES OF DESCRIPTIONS OF READING DIFFICULTIES

SELECTED BY TEACHERS IN THE CLASSROOM

 

 

 

 

 

TEACHERS
1==LA=I=3=LA=Q 6 7 8 9 l0 TOTAL—
MICHELLE x x x x x x x 7
JOE x x x x x 5
JIM x x x x x 5
DESCRIPTIONS
DIANE x x x 3

 

 

 

 

 

 

 

 

 

 

 

 

JOE: Corresponds to simulated case No. 7
JIM: Corresponds to simulated case No. 8

193

 

APPENDIX F

DIAGNOSTIC DOMAIN: CASE 7 and 8

N OHOg fl mm UHObv QUNUHH.

N cacao hum> "unequSufiumnsm vac:
.N xwua_huo> "mmumoumuum coauuuwmauamvw pups
.x .x .x x x ~3an 3335
.N on uswfia u“ wags um nomads» can whoa mo mafiaaawmn man an axooa
.N N3M£m mauuwa m "nouufinwouou vuo3
N No "nouumowufiuaopa who:
.N wmamm oauuHH mafixma mm? wuwvwmu cog: wawuouuuoo uaosuaa vamp ou «mudguaou
.N N mmvuaaoax Aaowumauossmv manna ”soaumSuuasm
N usuam>ouqaw ovum: "wdwvwou Homo
N Nausuuamon upoum
N HHoa.hum> apnea
N mwsfiummscpuos.ocwmov ou muoauwafimov mo kuumafi noaaamxm «on:
.N wouwawa on ou ammo unummoaoo
N .N sums. "Numaaawuo>
.N .N N munoocou gonna maogx ”muaooaou NHdHSbwoo>
wouafiun mu
N umga Na pmaamuumsoo on umsa mafia :Nuoum: may uMSu pnmumuopas uoa has mam
N oxma ou some you mmop ”maoauuundoo mosmusomuousH
N Nous "dogmausouaaoo u>aumunuouaH
N commas "huoum onu cw mucosuom vac uuowmo can mango mo mucosmam
.N .N .N noon “cowwamnauaaou
.N wawumua nuwa mvmuN
N No "Annouumuwao .wcwuufimv maqwuov Honouwa
.N .N maoaumwsv gnu mo meow mommaa am53.am>m .oumavava ”nowmaunounaoo

 

CH a w n o n q n N H

mmmmudmfi mBZmSUDDb UHBmozwfiHQ

h Mmdu ”2H4200 UHBmOzwfiHQ

h NHQmemd

194

195

0H

«3%
Q")
”Q”

MN
mm
FIN
H

CNN

.N

.N

.N

N

m m n o n e m N H

Ofi'

\Dfi'

\Dx'f
II

sooumnmao ca voaouunoa u .
Nuouuuonma aw vmsoauaoa I N

l [I 1‘

. . . . . usuavuuu uoa www.moaumas.vu03_aoasooa=
.3oaon haunwwan “Hm>mH macaw

muoauusum onww> o>wu

pawumumvns no: moov "muumawsfinm<

woman uuuauwv mum mamanoum no: mo maoauoumomasma may
msucmoa_nufia_waapmou ma one away uauumamwv mu amanoun no: «H
noon “aduuhm nowamam

mama ”Aoaumsosm saga umsumuv muouuo Huuaumaawuu

amanoua m we upoonuonswwms can «so: scum mosoaamca owmawawq
Noamuofimmv Honouaao uwmawsma a was umwmawama

meow "mofiwmuuuum nowuuouuou

xuoz_um«mhamau pups

wsanmamam coca hogan“ mono; um oommoau

maum> Naumasowuumm .mpuoa ou moanmaama muuxu mvn<
coauwowmausmvw whoa aw was ou uxmusoo mo on: vmuuaaa “uxuuaoo
sauna ou so: vaoumumuas you mmov "wawxnano

No "mono uxuuaou

No “wafixaanu

Nuoa.vuoa “maoao>

mo mamas mwwa on 0» atoms “mauso> use muamsomsoo
uamvaonmwaw on umwmhaaau owuosonm

axoma “mHHme ouumaonm

ozoax umvnsom wawasfiwum

Sosa «0 Name uumuuoo pap :muouum cowumuunaoaoummfiz

QQHOGwH :A:m._u.—: :Uflw: .1503: :09: Mu—HHV w—uuo3 HHGn—m
pmawwpwaom no: no: “Numaonaoo> unwwm

mono afiumaoanoun m "maumu0>mN

mpuoa unwwm canon so monomfia "aoaufiamooum

voow "humasnmoo> unwwm

uooamfip uumamwu mwuo3 mo paw on» ad monumfia «soaumouwausmvfi who:

196

.N
N
N
.N .N
.N
.N
.N
.N
N
N
N .N
N
.N N N
.N
N h
.N .N
N
N
N
.N N
.N
.N
CH o w n o m

mNWflUdflB

<r><>é><

.N oaoax "occaoo wawasuwmm

N oaoo onu uaono o>o£ odowuauwuonoo you moo: mam oouoa "oaoquuwuonoo who:

oona ”oxaao NOOH uonu mono:

oouo amanoua uoHooHo>ou vuoz

.N .N .N Nook. "mono: unwwm

No "soauoaooa aw who: unwwm

wdouuo “mono: unwfim

ooocoooaouoo ouoofipcfi ououuo "coauoowmauaooa who:

wouoaoma poo amuufiua coca uoosoouomawou poo xuoa Homo swaounu nouuon uou

opwo>o "mono: :Hs: no :3:

opwo>o nachos zq: Houuoq

N Ou aowusouuo Non uoc omov "nowuoouonam

mono: :30 won a: ooxoz

onaxo "opuoz

N Eouw mucouowooo no“: ocofiuwaamop puoa o>aw ou poo "odowuwafimov vuo3

mo odouuwuamov Nuouoowouuoo oo>ww ”ovuoa NQN

Nuoaonooo> ouoop o no: “Nuoaonooo>

.N muoovopo "ouaooooo whoa

Noo3 "uxouooo Nuouo mo mahou ca oooaouaom

Nooa "aoaoaosouuaoo HowuaouomsH

poom "oafiouop pom Nuoaoz

boom uwswwoou mo uoo wcfisooa wawuuow yam ooawououum

ouosvopo "aoaoconoumaoo Honouaa

mafiaouofia gog3 boom "wafiwdououovsp

poow "soamaosouaaou wawpoou Homo

.N .N .N ouosvovo "soaooosounaoo
m N H

mBZMSQDDb UHBmOZU<HD

m mmfiu "2Hdzoo UHBmOZO€HD

197

 

a m N o m e m N H

MHoouon usono ouoooouH

no: you NuuouHuaoo mo NooH ”aouwoun wquooN

oouo soHnoua "GOHuoHoaouu uoHum

Noou osu ou ouoodopo "ooHuHHHno Honuaoosoo
xuos.Hoo£oo waHuooaos on hos oaoHpoun HoHooo no Hosoauoam
muouo mo mocoo aufis ouomuousH ou ammo u.sov ououuo HoUHuooHoHn
Ho>oH ououw quSu uo SUHnoun owoowaoH on “owoowcoa

ou oHnosn "oHomHoao whoa aH ooHHHaom who? ououosou
mo: uoc ooov “who; azoaxa: osu nufis AHos ou uxousoo
oouo ENHaoum u:o= Hoswm

Sufi: oaanoua “wawxaano

Noo3 "oo>woooooom

xooz_ "owaHvao Houon

o>oa uos ooop "ooawououuo sowuoouuoo

oNooH ”opuoa uao «Human ou hwououum

oNooH ”oHHon Noouu<

ouoHumouaao NHHosuxoucoo ouo oaoo “maOHusuHuonso vuoz
wawaooa How Aoouo uxousOUV moo: oosfluoaoo "mono unoucoo
Ncano ou NuHHHno oaoo “ovHOB vaaonaoo

3on3 no: No8 "soHuoHooH :H oanosowuoHou Hoaaho vasom
aHon Saws pooHn pHsou "waHoooHn vaaom

no mo: uaouoHoaouaH "ovaon

uso canon ou unsouuo o: ”ovuoa uoHHHsowo:

saws msoHnoun "oucosooaou poo oHozo>

nufla oaanoua uwsfivaon

noon uoHHon coHuooHpoHHNm

Now: "ouaocooaou waHvao poo oHopfiz

Nook “oHoso> wdoH

Now? uoHozo> uuonm

Nova ”ocoHuoHooooo vcaoo wouuoH

Nova "oHHHNo oHconm

amHnoun o: “waHpcoHn fiasco Houa<

mo: ou NuHHHno oaoo ”waHnsoHo oHaoHHNm

wsouuo "oucoaooaoo HoHuHaH

osoax ”oaowuoaooooo venom uouuoH

198

 

 

N c e H N o N 0H m e u . soouoooHo ow voaowudoa I .

S a 2 N n 2 NH 3 m n u x 33333 38385 n N

N oHHHoooa “osoHnoun aOHumoouom
oH m .w B. o m a mu N H

APPENDIX G

LENGTH OF INTERACTION, NUMBER OF CUES COLLECTED AND NUMBER

OF DIAGNOSTIC JUDGMENTS FOR CASES 7 AND 8 AND 10 TEACHERS

APPENDIX G

NUMBER OF CUES COLLECTED AND NUMBER

LENGTH OF INTERACTION,

OF DIAGNOSTIC JUDGMENTS FOR CASES 7 and 8 AND 10 TEACHERS

 

oomocmoHp cow:

 

 

 

 

 

H.m o.m m.m m.m m.N o.m o.NH m.OH o.NH m.h m.m
o.m OH m HH N m MH NH vH m h m omoo
N.n m m m m m HH m CH m m h mmou
mmmocmoHo
mo Hmnaoz
m.m¢ o.Nm o.ov m.wv o.mH m.mm o.N> m.om m.Hm m.Nw m.hm mono coo:
v.Hm no mo mo NH mm mm on Nm an mm m omoo
m.N¢ Hw mm mo NH mm am no Hm we on N mmou
mono mo Honfisz
h.ov o.om m.hm m.nm m.mN o.Hv m.mm o.H¢ o.N¢ o.>¢ o.Nv mEHa coo:
H.mv om Ho om mN no hm wv mv No om m omou
m.H¢ ow vm mm mN mm om mm mm Nm «N h omou
Nmoa so oEHB
N 0H m m b o n v m N H
mammomma munamnadam

199

APPENDIX H

DIAGNOSTIC DOMAIN BASED ON CLASSROOM INTERVIEWS

 

N .N .N
N
N
N.N .N N N N
N N N N N
N .N N N N
.N N .N N
.N
N
N
N
N
N.N .N N N N .N .N .N .N
.N N N .N N N .N
N
.N .N .N
.N .N
.N N .N .N .N N
.N
.N
N .N
N
N .N N
N N N .N
.N N N N N .N N N
.N N N .N .N .N N
N .N N N N N . N .N
CH m m N o n e m N H
mmmmUde

ooooosu

ooaHo ououon

:o: omoun

85 En

waHvouoa

oHoNHoao vuoz

xoouuo vuox

Huoa_souuHH3 vao Hoao

mono: ounaocoun uonsoo
ovuoz_uoHHaHo oouo>oa
cowuoHooH dH mono:
huoHonooo> waHvooN
muoHanooo> uanm

coauwawooou pups

ouoouaoo mo wauauom

waacooa NuoHapouo>

axouaoo aH wdaaooa huoHsnooo>
NuoHonouo>

Honuowou cowuoauomsH waauuom
uxouaou mo: uoa moon

uxounou huoum

nacho GH aOHuoahomsH wsHuuow maoHnoum
aoHoconounsoo HoHuGouomaH poo Ho9uoom
aoHoconmunaoo waHaouoHH
GOHoconouaaoo wchoou usoHHm
aOHoconoumsou wfiwwoou Houo
:oaoaoaouasou

mmHmowmfido UHBmOZU<HQ

mSHH>mmBZH SOONmm<HU ZO ammdm ZHdSOQ UHBWOZUGHQ

m NHDZMHQN

‘200

201

 

.N .N N N
N N
N N N .N N N
.N
N N
.N
N
N N
.N N .N N
N
N
N
N N
N .N
.N N N N .N
N .N
.N .N
N
N .N .N N
N N N N
.N N
.N .N .N
.N .N
.N N N N .N
N N N .N N
.N N N .N N
N
N
N
N
N
N
N N N N
N
N
CH o m N o m Q m N H

soauoHoaouu uaHum
ooSHo axouaou
uxouaou

ovaaoo waHvGon
oaaouwown

ovnsoo HoHuHcH
ouuoaona nouuHuz
opaon

oHHon oHuoaonm

use oHoao> oo>omH
ovaaoo nouuoH

oooHu UHuoaonm
oHoso> waoH

cocoon Hoaob

oHoso> uuonm

oHoao>
monopaoaoouuoo AoHsoaonmonaouwv vanoo Houuoa
ousoaooaoo HoHuHaH
uao bosom ou oHA<
suooummo AHfiaom who:
opsooo usosooooo
ovaaoo wawuam

oucaoo ovafiz

ovasoo waHadeom
ovasom

oHoNHoao oHuoaonm
ooNmeom

oowaoum

ouauoauuo ofismaonmonmuoz
oHoHSHm
ovuoa.vaoonsoo
mqumHHNm

waaxaonu

oHoNHoao Houauoauum
ouuon any coca nonuou who: oHoaa onu ovum

202

HH 0 w m m m NH mH N N I .. knouoHOHoH aw Goaunma I .
0N NH 0N NN eN mH «N NN NH on I N aoonoooHu ca pmfioauaua I N

N ononosuH

N pononnoo MHom

N vonoonuowv NHHoom

N oOHoH> nonoa noom

n aanona Huamn>

N osoHnona Hoauoounom

N unopano vnoaon ovoann<

N oamHHonm HooHoNsm

N xnos_aonuwn3

N xnoz.owoswdoa

.N N naooooo HHom

.N vo>Hnnov NHHonouHou

N ooHnoHawnHH

N Nnoaua Hono

N aOHuoaHonooo none:

N nooomm

N oanon wnHNnoa noon

x Haswannnm

.N .N nooHoHn

opasoo Nnouava<

aoHnonm Nnosuz

sownnonno wdHNoa noz

munnnnumnu mananomn

Nnoaua Hoaow>

Nnosoa.NnonHv:<

onouoom o.onno

.N Ho>oH ovonw SOHon waaooom

.N N aOHuoanunom

.N .N ooxonoua no: onoonnoo

.N oonoa.omwxm

N ooaoofiz

N ooooo vnoz

N N oaoauauanopam

.N wagooon unonHm

N N N N woavoon Hono
o c m

NNNNNN

HN N
mN
mN
h~>< ><

APPENDIX I

OBSERVATIONAL STUDY DATA ANALYSIS

SYSTEM (OSDAS) STATISTICS

APPENDIX I

OBSERVATIONAL STUDY DATA ANALYSIS

SYSTEM (OSDAS) STATISTICS

‘ The following statistics are all calculated by (l)
a computer statistical analysis system (Observational Study
Data Analysis System) developed and maintained by the
Institute for Research on Teaching, Michigan State
University, and (2) an interactive statistical analysis
system (MIDAS) at the University of Michigan. A random
sampling of data was computed by hand across numerous

observational studies and the system was found to be operating

accurately.

PROPORTIONAL AGREEMENT

Given a domain for diagnoses/remediations/ or
cues (DX/Rx/ or CX) for a given case, proportional agreement
is the proportion of clinicians who mentioned each Dx/Rx or
CX statement. One proportion is computed for each statement
in the domain.

The statistic is bounded by 0 and l and is calculated
by dividing the number of clinicians who mentioned a statement

by the total number of clinicians.

203

204

number of

Cji where Cji

Cj clinicians
mentioning the
ij category

Cj = total number
of clinicians
for a given
case.
For example, if 3 clinicians of a total of 6 who

diagnosed a given case mentioned the category "Basic sight

words, weakness", the proportional agreement would be:

No. of clinicians mentioning the category

3
POA. = —— = 050
6
COMMONALITY

Given a domain for diagnoses/remediations/ or cues
(DX/RX/CX) for a given case, the commonality statistic is a
measure of agreement between one clinical session and all
other clinical sessions for a given case, e.g., an
individual is being compared with a group.

The statistic is bounded by 0 and l. A value of x
for a given clinician roughly implies that she/he has

included in her/his session x percent of those statements

205

mentioned most frequently by the group for that case.

For example, a clinician who has a score of .34

has included in her/his session roughly 34 percent of those

statements mentioned most frequently by the group.

INTER-CLINICIAN CORRELATION

Given a domain for diagnoses/remediations/ or cues

(DX/Rx or CX) for a given case, the phi correlation is a

measure of inter-clinician agreement.

computed for each pair of clinicians.

One phi correlation is

The DX/RX/CX categories mentioned by one clinician

are compared with those mentioned by a second clinician for

the same case.

below.

CLINICIAN B,

SIMCASE Y

62mmmwo

HZMUIUI'JP'

This comparison is summarized in the table

(+)

(-)

CLINICIAN A,

PRESENT (+)

SIMCASE Y

ABSENT (-)

 

Frequency count of
statements in the
domain present in
both clinicians
DX/RX or CX

Frequency count of
statements in the
domain present in
clinician B's
session but not in
clinician A's
DX/RX/ or CX

 

 

Frequency count of
statements in the
domain present in
clinician A's session
but not in B's DX/RX
or CX

 

Frequency count of
statements in the
domain absent in
both clinicians'
DX/RX or CX

 

 

206

The calculation of the Phi correlation is derived
from the table as follows:

Clinician A, Simcase Y

 

 

 

 

 

 

+ -
+ a(++) b(+-) a+b
Clinician B, _
Simcase Y
- c(-+) d(--) c+d
a+c

I
b+d | N

I

I

Phi = (axd - bxc)

 

(a+c)x(b+d)x(c+d)x(a+b)

The statistic is bounded by -l (statements are in
cells b and c only) and l (statements are in cells a and d
only), only if the distribution in the marginals are equal.
In all other cases the maximum and minimum values will be
less than 1 and greater than -1.

An example of a completed table is as follows:

Statements of Statements of Domain of
Clinician A, Clinician B, statements
Simcase Y Simcase Y
$1 $1 81
52 $2 82
S3 S7 S3
S4
SS
36

S7

207

Clinician A

 

 

 

 

 

+ —
+
2 l 3
Clinician B
- l 3 4
l
3 I 4 7
l
u

 

LI ST OF REFERENCES

LI ST OF REFERENCES

Adams, R.K. Early identification of potentially retarded
readers within the school system. In E.O. Calkins
(Ed.), Readinngorum (NINDS Monograph no. 11).
Bethesda, Md: U.S. Department of Health, Education
and Welfare, 1971.

 

Austin, M.C. Diagnostic teaching: The primary grades.
In J.H. Root (Ed.), Diagnostic teaching: Methods
and materials. Syracuse, N.Y.: Syracuse University,
School of Education, 1965.

 

 

Austin, M.C. Identifying readers who need corrective
instruction. In D.G. Schubert & T.L. Targerson (Eds.),
Readings in reading: Practice theory research.

New York; Thomas Y. Crowell Company, 1968.

 

Bamman, H.A. Assessing progress in reading. In R. Farr
(Ed.), Measurement and evaluation of reading.
New York: Harcourt, Brace and World, Inc. 1970.

 

Barr, R. & Sadow, R. Classroom reading diagnosis and
instructional planning. Unpublished manuscript,
University of Chicago, 1978.

 

 

Bateman, B. The role of individual diagnosis in remedial
planning for reading disorders. In E.O. Calkins
(Ed.), Readinngorum (NINDS Monograph no. 11).
Bethesda, Md: U.S. Department of Health,
Education and Welfare, 1971.

 

Blount, D. Competencies essential for diagnosing reading
difficulties. Teacher Corps Association: Resources
for CBTE, no. 6, 1973. ED 095 151.

 

 

Bond, G.L. Diagnostic teaching in the classroom. In D.L.
De Boer (Ed.), Reading diagnosis and evaluation.
Proceedings of the thirteenth annual convention of
the International Reading Association, (Vol. 13),
Newark, Del: International Reading Association,
1970.

 

Bond, G.L. & Tinker, M.A. Reading difficulties: Their
diagnosis and correction. (2nd ed.). New York:
Meredith Publishing Company, 1967.

 

 

208

209

Buerger, T.A. A follow up of remedial reading instruction.
Reading Teacher, 1968, 21, 329-334.

 

Burnett, R.W. The classroom teacher as a diagnostician. In
D.L. De Boer (Ed.), Reading diagnosis and evaluation.
Proceedings of the thirteenth annual convention of
the International Reading Association (Vol. 13).
Newark, Del: International Reading Association,
1970.

 

Burnett, R.W. The diagnostic proficiency of teachers of
reading. The Reading Teacher, 1963, 16 (4), 229-34.

 

Carter, H.L.J. The teacher's diagnosis of the disabled
reader. In W.K. Durr (Ed.), Reading difficulties:
Diagnosis, correction and remediation. Newark,
Del: International Reading Association, 1970.

 

 

Carter, H.L.J. & McGinnis, D.J. Diagnosis and treatment of
the disabled reader. London: The MacMillan Company,
1970.

 

 

Cartwright, G.P., Cartwright, C.A., & Yssledyke, J.E. Two
decision models: Identification and diagnostic
teaching of handicapped children in the regular
classroom. Psychology in the Schools, 1973, 19,
4-11.

 

Dauzat, S.W. Informal diagnosis: The nucleus. In R.A.
Earle (Ed.), Classroom practice in reading. Newark,
Del: International Reading Association, 1977.

 

Della-Piana, G.M. Reading diagnosis and prescriptions:
An introduction. New York: Holt, Rinehart and
Winston, Inciil968.

 

 

Dietrich, D.M. A diagnostic approach to corrective reading
in the classroom. (ERIC Document Reproduction Service
No. 063 576). Paper presented at the annual meeting
of the International Reading Association, Detroit,
May 1972.

 

 

Dupuis, M.M. & Ashkov, E.N. Diagnostic teaching in content
area reading. In P.D. Pearson (Ed.), Reading: Theory,
research and practice. Twenty-sixth yearbook of the
National Reading Conference. St. Paul, Minn: Mason
Publishing Company, 1977.

 

 

Ekwall, E.E. Diagnosis and remediation of the disabled
reader. Boston, Mass: Allyn and Bacon, Inc.'1976.

 

Ekwall, E.E. Teacher's handbook on diagnosis and remediation
in reading. Boston, Mass: Allyn and Bacon, Inc.
1977.

210

Elstein, A.S., Shulman, L.S., & Sprafka, S. Medical problem
solving: An analysis of clinical reasoning.
Cambridge, Mass: Harvard University Press, 1978.

Farr, R. Reading diagnosis: Trends and issues. (ERIC
Document Reproduction Service No. 050 899). Paper
presented at the meeting of the Nassau Reading
Council, Uniondale, New York, March 1971.

Florida Department of Education, Division of Public Schools,
Bureau of Curriculum Services. Diagnostic
prescriptive reading instruction. (ERIC Document
Reproduction SerVice No. 149 310), 1977.

Gallant, R. The development of a diagnostic instrument.
In R. Farr (Ed.), Measurement and evaluation of
reading. New York: Harcourt, Brace and World, Inc.,
1970.

Garland, L.H. Studies on the accuracy of diagnostic pro-
cedures. American Journal of Roentgenology,
1959, 83, 25-38.

Gil, D., Hoffmeyer, E.C.M., Van Roekel, J.L. & Weinshank,
A.B. Clinical problem solving in reading: Theory
and research. (Res. Ser. No. 45). East LanSing,
Mich: Institute for Research on Teaching, Michigan
State University, 1979.

 

Gil, D., Vinsonhaler, J.F., & Wagner, C.C. Studies of
clinical problem solving behavior in reading diagnosis
(Res. Ser. No. 42). East Lansing, Mich: Institute
for Research on Teaching, Michigan State University,
1979.

 

Gil, D., Wagner, C.C. & Vinsonhaler, J.F. Simulating the
clinical problem solving behavior of reading
clinicians (Res. Ser. No. 30). East Lansing, Mich:
Institute for Research on Teaching, Michigan State
University, 1979.

 

Glad, J.R. Evaluation of the remedial reading program in
Utah Public Schools. UnpubliSHed doctoral diSsertation,
Salt Lake City, Utah: University of Utah, 1965.

 

Guszak, F.J. Diagnostic reading instruction intthe ele-
mentary school. New York: Harper and Row Publishers,
I972.

 

211

Harris, A. J. Can the teacher of reading be a dia nostician?
(ERIC Document Reproductionfi Service No. 032 909).
Paper presented at the annual meeting of the
International Reading Association, New Orleans,
May 1974.

Harris, A.J. Diagnosis of reading disabilities. In H. M.
Robinson (Ed. ), Corrective reading in classroom and
clinic. Supplementary Edfibational Monographs No. 70
Chicago: University of Chicago Press, 1953.

Harris, A. J. The diagnosis of reading disabilities. In
A. J. Harris 8 E. R. Sipay (Eds. ), ReadingsJ on reading
instruction (2nd ed. ). New York: David McKay

 

 

Harris, A. J. The reading teacher as diagnostician. In R.A.
Earle (Ed. ), Classroom practice in reading. Newark,
Del: International Reading Association, 1977.

Harris, A.J. & Sipay, E. R. How to increase freading_abiligy.
(6th ed.). New York: David McKay Company, Inc., 1975.

 

Harris, A.J. & Smith, C.B. Dingnostic teaching in the
classroom. New York: Holt, Reinhart and Winston,
1976.

Hollingsworth, P. M. Diagnosis and pro nosis: An
interdisciplinary approach. (ERIC Document Repro-
duction Service No. 042 582). Paper presented at

the annual meeting of the International Reading
Association, Anaheim, Calif., May 1970.

 

 

Jan-Tausch, E. Reading diagnosis: The essential ingredient
in teaching every child to read. In R.E. Leibert
(Ed.), Diagnostic viewpoints in reading. Newark,
Del: International Reading Association, 1971.

Karlsen, B. Reading: Assessment and diagnosis of abilities.
In P. Lamb & R. Arnold (Eds. ), Reading: Foundations
and instrugnional strategies. Belmot, Ca1if.:
Wadsworth Publishing Company, Inc., 1976.

Kennedy, E.C. Classroom a roaches to remedial reading.
Itasca, III.: F.E. Peacock PubliShing, Inc., 1971.

Kress, R.A. Diagnosis: An interdisciplinary approach. In
J.A. Figurel (Ed.), Reading and inguiry. Conference

proceedings (Vol. 10). NewarE, De : International
Reading Association, 1965.

 

 

 

McCarthy, W.G. Diagnosis, prescription, correction and
evaluation in the teaching of reading. Contemporany
Education, 1971, 42 (5), 233-235.

 

 

212

 

McGinnis, D.J. The teacher's approach to the stud of the
disabled reader, A homeopathic concept. (ERIC
Document Reproduction Service No. 042 577). Paper
presented at the annual meeting of the International
Reading Association, Anaheim, Calif., May 1970.

Medrano, P.P. A psychatric perspective of reading disability.
In W. Otto, N.A. Peters & C.W. Peters (Eds.),
Readingpproblems: A multidisciplinary perspective.
Cambridge, Mass.: Addison-Wesley Publishing Company,
1977.

Monroe, M. Children who cannot read. Chicago: University
of Chicago Press, 1932.

 

Monroe, M. General principles of diagnosis of reading
disabilities. In D.G. Schubert & T.L. Targerson
(Eds.), Readings in reading: Practice theory research.
New York: Thomas Y. Crowell Company, 1968.

Niles, O.S. Diagnostic reaching in the junior high school.
In J.H. Root (ed.), Diagnostig teaching: Methods and
materials. Syracuse, N.Y.: Syracuse University,
School of Education, 1965.

 

Olson, J.P. & Dillner, M.H. Learning to teach reading in the
elementary school utilizing a competency based
instructional nystem. New York: Macmillan Publishing
Co., Inc. 1976.

Otto, W., McMenemy, R.A. & Smith, R.J. Corrective and
remedial teaching (2nd ed.). Boston: Houghton Mifflin
Company, 1973.

 

Paton, B.C. The accuracy of diagnosis of myocardial
infarction. American Journal of Medicine,
1957, 23, 761-768.

Pearson, P.D., & Kamil, M.L. Basicpprocesses and
instructional practices in teaching reading (Reading
Education Report No. 7YT’ Urbana-Champaign, Ill:
Center for the Study of Reading, University of
Illinois at Urbana-Champaign, 1978.

Peters, C.W. Diagnosis of reading problems. In W. Otto.,
N.A. Peters, & C.W. Peters (Eds.), Reading problems:

A multidisciplinar ers ective. Cambridge, Mass.:
Addison-WesIey PuEIishing Company, 1977.

 

213

Peters, N.A. An interdisciplinary approach to the
assessment and management of severe language-
learning problems. In W. Otto., N.A. Peters. &
C.W. Peters (Eds.), Reading‘pnoblems: A multi-

disciplinary pers ective. CamBridge, Mass.:
Addison-Wesley PuEIishing Company, 1977.

 

Rabinovitch, R. Differential diagnosis in children with
reading retardation. In J.H. Root (Ed.),
Diagnositic teaching: Methods and materials.
Syracuse, N.Y.: Syracuse University, School of
Education, 1965.

Reich, R.R. More than remedial reading. In A.J. Harris &
E.R. Sipay (Eds.), Readings in readinginstrnction
(2nd ed.). New York: David McKay Company, Inc.,
1972.

Robinson, H.M. Why pnpils failin reading. Chicago:
University of Chicago Press, 1946.

 

Satz, P. Reading problems in perspective. In W. Otto.,
N.A. Peters, & C.W. Peters (Eds.), Reading problems:
A multidisclplinary pergpective. Cambridge, Mass.:
Addison-Wesley Publishing Compnay, 1977.

 

Sawyer, D.J. The diagnostic mystique - a point of view.
The Reading Teacher, 1974, 27 (6), 555-561.

Sawyer, R. Diagnosis for the classroom teacher (ERIC
Document Reproductidn Service No. 029 754). Paper
presented at the conference of the College Reading
Association, Knoxville, April 1968.

Sheldon, W.D. Specific principles essential to classroom
diagnosis. In D.G. Schubert & T.L. Targerson
(EDs.), Readings in rnnding:iPractice theory research.
New York: Thomas Y. Crowell Company, 1968.

 

Sheldon, W.D. Teachers in grades K-12 diagnose reading
problems. In J.H. Root (Ed.), Diagnostic teaching:
Methods and materials. Syracuse, N.Y.: Syracuse
University, School of Education, 1965.

 

 

Sherman, G. A model of rgadingand leanning to read. East
Lansing, Mich.: Institute for Research on Teaching,
Michigan State University, in preparation.

Sherman, G. Training reading specialists in diagnosis (Res.
Ser. No.931). East Lansing, Mich.: Institute for
Research on Teaching, Michigan State University, 1978.

214

Sherman, M. Psychiatric insights into reading problems. In
D.G. Schubert & T.L. Targerson (Eds.), Readings in
reading: Practice theory research. New York:
Thomas Y. Crowell Company, 1968?

 

Smith, C.B. Correcting readingproblnns in the classroom.
Newark, Del.: International Reading Associatién, 1969.’

Smith, C.B., Carter, B., & Dapper, G. Treating reading_
difficulties: The role of the principal, teacher,
npecialiét, administrator. Washington, D.C.: U.S.
Deaprtment of Health, Education, and Welfare, 1970.

 

Smith, F. Understanding reading: A psycholinguistic
analysis of reading and learning to read. New York:
Holt. Reinhart and'Winston, Inc., 1971.

Spache, G.D. Clinical diagnosis in the classroom. The
Reading Teacher, 1960, ll, 14-18.

 

Spache, G.D. Diagnosing and correcting reading disabilities.
Boston: Allyn and Bacon, Inc.,—1976.

Spache, G.D. Diagnosis of reading problems. The Elementary
School Journal, 1955, 56, 18-26.

 

 

Stephens, E.L.A. A study of the_prob1em solving behavior
of teachers as they diagnose a childTS reading
performance and’the effects 6f experience and training
on that behavior. Unpuhlished doctoral dissertaEion,
Michigan State University, 1978.

 

Stevens, D.O. Reading difficulty and classroom acceptance.
The Reading Teacher, 1971, g; (1), 52-55.

Strang, R. Dia nostic teachin of reading. New York:
McGraw-HII1 Company, 1 42

Strang, R. Levels of reading diagnosis. Educational Forum,h

 

Strang, R. Reading diagnosis and remediation. Newark,
Del.: International Reading Association, 1968.

Swalm. J.E. Diagnostic teaching: A method for assessing
reading skills. In M. Kling (Ed.), Language develnp-
ment for the classroom and renedial reading.
Proceedings of the fall 1972 Rutgers University
Reading Confernece, New Brunswick, 1973. (ERIC
Document Reproduction Serivce No. 140 260).

 

 

215

Van Roekel, J. L. 8 Patriarca, L. Do reading and learning
disabilityrclinicians differ in Eheir dIagnosis and
remediation? Paper presented’at the annual meeting of
the International Reading Association, Atlanta,

April 1979.

 

 

 

Vinsonhaler, J.P. Clinical problem solving among reading
specialists: The problem of agreement on a diagnosis.
Paper presented at the annuai meeting of the
International Reading Association, Atlanta, April
1979.

 

Vinsonhaler, J.F., Wagner, C.C. & Elstein, A.S. The Inquiry
Theory: An information processing approach to
clinical problem solving research and application.
Medinfo '77. Proceedings of the Second World

Conference in Medical Information, 1977.

Wagner, C.C., Kitchen, F., Polin, R., Vinsonhaler, J.E.,
Van Roekel, J.L. & Gil, D. gngervational studies:
Data analysis system. (Tech. Rep). ~East Lansing,
Mich.: Institute for Research on Teaching, Michigan
State University, 1979. '

Weule, G.G. Diagnosis and treatment of reading problems:
The teacher's role. The Slow Learning.Child, 1971,
.l§ (3), 168-188.

Weinshank, A.B. An observational study of the association
between diagnosis and remediation in reading. Ph.D.
dissertation, Michigan State UniVersiEy, in progress.

Weinshank, A.B. The relationships between diagnosis and
remediation ih reading: A pilot studyn (Res. Ser.
No. 37): East Lansing, Mich.: Institute for
Research on Teaching, Michigan State University,
1979.

Weintraub, S., & Farr, R. Introduction. In R. Farr, S.
Weintraub & B. Tone (Eds.), Inproving reading research.
Newark, Del.: International Reading Association,
1976.

Whitcombe, L.M. Overcoming reading difficulties and the
effect on personal development. In R.T. Williams
(Ed. ), Insights into why and how to read. Newark,
Del.: International Reading Association, 1976.

Wilson, R.M. Diagnostic and remedial reading for classroom
and clinic (3rd ed. ). Cqumbus, Ohio: Charles? E.
Merrill Publishing Company, 1977.

 

216

Yerushalmy, J. Reliability of chest radiography in the
diagnosis of pulmonary lesions. American Journal of
Surgery, 1955, 89, 231-240.