University of Southern California Dissertations and Theses

Using task based writing instruction to provide differentiated instruction for English language learners

(USC Thesis Other)

Using task based writing instruction to provide differentiated instruction for English language learners

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content USING TASK BASED WRITING INSTRUCTION TO PROVIDE
DIFFERENTIATED INSTRUCTION FOR ENGLISH LANGUAGE LEARNERS
by
Alexandros Merkouris Bantis
____________________________________________________________________
A Thesis Presented to the
FACULTY OF THE ROSSIER SCHOOL OF EDUCATION
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfilment of the
Requirements for the Degree
MASTER OF SCIENCE
(TEACHING ENGLISH TO SPEAKERS OF OTHER LANGUAGES)
May 2008
Copyright 2008 Alexandros Merkouris Bantis
EPIGRAPH
Written speech is monologous; it is a conversation with a blank sheet of
paper. Thus writing requires a double abstraction: abstraction from the sound
of speech and abstraction from the interlocutor ... our studies show that the
child has little motivation to learn writing when we begin to teach it. He feels
no need for it and has only a vague idea of its usefulness.
--Lev Vygotsky, Thought & Language, 1986, p.181
ii
DEDICATION
What I remember most about Christine were her large brown eyes that
seemed to sprout out of her rosy round face. She was in the second grade and had
started learning English just a few years before. One day her teacher came to me for
help because she was throwing chairs in the classroom. I sat down in a chair too
small beside that little girl and gave her all my attention.
“Why?”
“Writing is hard.”
“Tell me, who reads what you write?”
“Nobody.”
This thesis is dedicated to the minority language children who throw chairs in
the classroom because nobody has taught them how to have a conversation with a
blank piece of paper.
iii
ACKNOWLEDGMENTS
I wish to express my deepest gratitude to the many people who made this
thesis possible but especially to three women.
First I wish to thank Dr. Eskey, who encouraged me to do a bona fide
research project and write a thesis. I always felt a tremendous sense of peace and
calm in her classroom. She planted the seed.
Second I wish to thank Dr. Clemons, who worked with me through the
formative stages of this thesis project. Of all my teachers at USC, I worked the
hardest for her because she has a passion for learning and compassion for the learner.
She watered the seed, and it sprouted and started growing.
Last and most of all, I wish to thank Ana Cecilia. She kept me company when
I studied through the night at Leavy Library. She continuously encouraged me to
finish what I had begun. Without her optimism and support, the seed that turned into
a young plant might have withered away. Her sunshine enabled it to bear fruit.
iv
TABLE OF CONTENTS
Epigraph ii
Dedication iii
Acknowledgments iv
List of Figures vii
Abbreviations viii
Abstract ix
Chapter 1: Introduction
Background 1
Problem Statement 2
Task Based Writing Instruction 3
Research Questions 4
Significance of the Study 5
Organization 6
Definitions 7
Limitations 10
Chapter 2: Literature Review
Introduction 11
California’s Education Crisis 12
Changes in Education Policy 14
The Research on Open Court Reading 16
Direct Instruction 19
Constructivist Pedagogy 24
Second Language Acquisition 28
Task-Based Instruction 29
Form-Focused Instruction 30
Task-Based Writing Instruction 35
Conclusion 38
Chapter 3: Method
Introduction 40
Research Questions 42
Research Context 43
Participants 51
Recruitment 53
v
Chapter 3: Method (Continued)
Task-Based Writing Instruction 55
Targeted Linguistic Features 59
Procedure 61
Reliability and Validity 69
Chapter 4: Results
Introduction 71
Primary Data Set 73
Quantitative Data 74
Qualitative Data 80
Conclusion 92
Chapter 5: Discussion
Introduction 95
Q1, Using TBwI in the Classroom 96
Q2, TBwI and Differentiated Instruction 101
Q3, TBwI and Second Language Acquisition 102
Recommendations for Practitioners 102
Policy Implications 104
Research Needed 107
Conclusion 108
Glossary 110
References 114
Appendices
List of Appendices 120
Appendix A: Participant Summary Data 121
Appendix B: Writing Prompts 131
Appendix C: Sample Writing Conference 133
Appendix D: Recruitment Tools 137
Appendix E: Development of Criteria for Data Coding 140
vi
LIST OF FIGURES
Figure 1: Sample Turns within Data Set 73
Figure 2: Overall Linguistic Focus of Prompts/Recasts 74
Figure 3: Ratio of Prompts to Recasts 76
Figure 4: Ratio of Turns to Sentences 77
Figure 5: Ratio of Teacher to Student Talk 79
Figure 6: Participant A Linguistic Focus 121
Figure 7: Participant B Linguistic Focus 122
Figure 8: Participant C Linguistic Focus 123
Figure 9: Participant D Linguistic Focus 124
Figure 10: Participant E Linguistic Focus 126
Figure 11: Participant F Linguistic Focus 127
Figure 12: Participant G Linguistic Focus 128
Figure 13: Participant H Linguistic Focus 129
Figure 14 : Participant I Linguistic Focus 130
vii
ABBREVIATIONS
ELLS: English Language Learners
ERIC: Education Resources Information Center
FFI: Form Focused Instruction
IWT: Independent Work Time
LAUSD: Los Angeles Unified School District
NNS: Non-native Speaker
NS: Native Speaker
OCR: Open Court Reading
TBI: Task Based Instruction
TBwI: Task Based Writing Instruction
TESOL: Teaching English to Speakers of Other Languages
ZPD: Zone of Proximal Development
viii
ABSTRACT
The purpose of this study was to investigate the impact of task based writing
instruction (TBwI) on English language acquisition and differentiated instruction for
minority language students during the Independent Work Time instructional
component of the Open Court Reading program. One teacher and 10 third grade
students (8-9 years old) participated in this mixed methods study. TBwI was as a
platform for communicative language teaching. Together they recast the students’
written interlanguage embedded into standard English. The study took place after
school, 45 minutes per day for one month, resulting in 35 transcribed writing
conferences, writing samples, and interviews. Results indicate that TBwI can be a
useful vehicle for differentiated instruction, constructivist pedagogy, and principles
of second language acquisition to address the diverse needs of second language
learners.
ix
CHAPTER 1, INTRODUCTION
Background
Many years ago I chose to enter the teaching profession because I wished to
be a catalyst for learning and knowledge in the lives of the children I was going to
come into contact with. I felt impelled to enter into a masters program in part from
my experience as a first year teacher. Those third grade children were learning
English as a second language and over half of them would eventually fail to join the
literacy club. They began the prior year in a bilingual education class, but when they
returned to class after Christmas break, the Spanish books were gone and replaced
with English-only instruction. Angie, one of those students, began that year with
strength and self-confidence, but one day she cried. She was forgetting how to write
in her mother tongue; she was also struggling to learn English as a second language.
Guided by these memories, I felt a sense of urgency during my masters program.
What could I have done differently to steer her clear of failure?
Angie is not alone. There are about 1.5 million minority language students in
California’s public schools, and many of them struggle to successfully complete high
school. High school is the end of the road for the educational aspirations of many
minority language learner students of California, a road that often begins when they
enter into a kindergarten class. Kindergarten children such as Angie speak little to no
English. They are confronted with a teacher using a curriculum designed for native
English speaking children, a curriculum that emphasizes direct instruction and
behaviorist pedagogy.
1
The literature of constructivist pedagogy contains a wealth of information
that could benefit Angie and her fellow minority language classmates. Teachers
could harness the potential of constructivist concepts such as zone of proximal
development (ZPD) and assimilation to more effectively guide minority language
students towards English language proficiency.
The literature of second language acquisition also contains a wealth of
information that could benefit Angie and her fellow minority language classmates. A
branch of communicative language teaching called task based instruction (TBI) as
well as form focused instruction (FFI) holds much promise. One variant of TBI/FFI
is called a jigsaw task (a two-way information-gap task), which ideally pairs a native
English speaker and an English language learner with an authentic communicative
task and problem to be solved. Using English as an authentic tool for communication
along with efficient techniques such as prompts and recasts, the native English
speaker could be a rich source of input and provide highly differentiated instruction.
Prompts can be an effective vehicle for injecting constructivist pedagogy into the
classroom.
Problem Statement
For many of the 1.5 million minority language students currently enrolled in
California’s public education system, there exists a mismatch between their second
language acquisition needs and the type of instruction they receive. Many receive
English only instruction using materials designed for native English speakers. Many
receive instruction using a pedagogy that is behaviorist in origin. Many learn in
2
classrooms that lack sufficient differentiation of instruction. Due in part to a failure
to achieve cognitive English language proficiency, many of these minority language
learners ultimately drop out of school.
The second language acquisition process is lengthy and complex often
fraught with failure even under ideal learning conditions. Because of lack of
differentiation and constructivist pedagogy, because of the absence of principles and
practices from the field of second language acquisition, the learning conditions for
these minority language learners are less than ideal.
Task Based Writing Instruction
One variant of the jigsaw task (a two-way information-gap task within task
based instruction) developed in conjunction with this study is task based writing
instruction (TBwI), which uses the writing conference as a vehicle for second
language acquisition.
Maintaining primacy of meaning, TBwI uses the writing conference as a
platform for authentic language use between the minority language student and the
teacher. The two work together to recast the student’s interlanguage (student
emergent abilities in the second language) embedded within a rough draft into
standard English. This could allow for highly differentiated language instruction.
The first step (pre-conference) requires the language learner to create a
written rough draft. The written piece needs to serve an authentic communicative
intent, such as making a response to literature or text-self connection. In the present
study, the language learners were prompted to write a single paragraph text-self
3
connection after reading an literature excerpt. They were given a choice of three
possible topic sentences; then they were prompted to write a rough draft paragraph
by choosing one of them and writing four supporting sentences to create a five
sentence paragraph.
The second step is a writing conference; the language learner (student) and
native-speaker (teacher) sit down together to discuss the rough draft. They discuss it
one sentence at a time. The teacher uses conversational techniques (prompts and
recasts) so that they may rewrite it into standard English. The language learner has
experiential expertise while the teacher has linguistic expertise. There is a focus on
meaning with an occasional shift to grammar. At the conclusion of the writing
conference, the student leaves with both the original rough draft and the final draft.
The third step (post-conference) occurs when the student compares the final
draft and rough draft so as to notice the changes made. This may consist in marking
up the rough draft with editing marks to reflect the changes. Finally, the student
writes the final draft in his/her own hand.
Research Questions
There were three research questions of the current study:
(1) What are the issues associated with using TBwI as a platform for
communicative language teaching consistent with principles of TBI/FFI and
constructivist pedagogy?
(2) How does TBwI impact the degree of differentiated instruction within a
mixed ability classroom?
4
(3) How does TBwI impact second language acquisition?
The overall goal of this study was to explore the use of TBwI for third grade
English language learners receiving instruction with Open Court Reading (OCR), an
English language arts program designed for native English speakers.
Significance of the Study
Developing a new tool (TBwI) to provide differentiated instruction,
constructivist pedagogy, and second language acquisition pedagogy has importance
from a theoretical perspective. It represents a new and potentially fruitful line of
investigation in communicative language teaching, and it also has practical
importance.
There exists a mismatch between the linguistic needs of many minority
language students and their English language arts curriculum. Some students may
respond positively to communicative language teaching approaches rather than
reading programs designed for native English speakers.
Within OCR, independent work time (IWT) is an instructional block of 20-40
minutes per day during which time students work in small groups or independently
on projects; meanwhile, the teacher pulls aside a flexible group of students or works
with them individually to provide differentiated instruction. It is during this time that
the teacher often will work with small groups of students on their writing, either
preteaching or reteaching OCR lessons.
This study pilots the application of TBwI within IWT of OCR to understand
the impact of diversifying the instructional approaches for those students.
5
Organization
This thesis contains five chapters.
Chapter two first examines the historical context and current state of public
education for English language learners in California. This section also examines the
research base both favorable to and critical of behaviorist pedagogy and OCR. Next,
this section examines the literature on select topics of second language acquisition
and constructivist pedagogy that may support English language learning.
Specifically, the section examines task based instruction, form focused instruction,
zone of proximal development, and the distinction within constructivist pedagogy
between accommodation and assimilation. Finally, it provides a description and
rationale for task-based writing instruction.
Chapter three describes the methodology used in the study. It establishes the
research questions and the process used to answer them. Background information
regarding characteristics of the school site, participants, and myself as participant-
researcher are provided. Various aspects of OCR that deal with writing instruction
and IWT are examined. Finally, the procedure for data collection (how the study was
conducted) and analysis (how the information was coded) are discussed.
Chapter four presents both quantitative and qualitative results of the study.
After the introduction, the primary data set is introduced, which was a synthesis of
the student rough drafts and transcriptions of the 35 writing conferences, broken
down to the conversational turn level, each turn coded according to targeted
linguistic feature and conversational technique (e.g., prompt, recast) used. Next,
6
information from quantitative data sources is presented and discussed. Finally,
information from qualitative data sources (the primary data set plus interviews) are
presented and organized according to the research questions and topics associated
with the research questions.
Chapter five discusses the results presented in the previous chapter, organized
around the three research questions. I have provided specific recommendations for
teacher practitioners who are interested in incorporating TBwI into their classroom. I
have also examined the broad policy implications of the study. Before concluding the
chapter with final thoughts, I have provided recommendations for additional research
needed.
The appendices provide a wealth of additional information. Appendix A
presents summary data for each individual student. Appendix B presents sample
writing prompts used for the study. Appendix C presents the detail for one sample
writing conference along with the student draft. Appendix D lists the recruitment
tools used. Finally, Appendix E recounts the iterative process used to arrive at the
final coding system.
Definitions
The following are select definitions that may provide clarification regarding
technical terms and concepts. Please consult the Glossary for additional terms
Behaviorist Pedagogy: A teaching method that emphasizes the ability to learn
new information through repetition and structured practice.
7
Constructivist Pedagogy: A teaching method that emphasizes the need for a
learner to actively participate in the learning process by incorporating new
information into previously internalized knowledge (De Lisi & Golbeck, 1999).
Direct Instruction Pedagogy: A teaching method that emphasizes fast-paced
and explicit instruction that may be highly scripted O'Neill (1988).
Differentiated Instruction: Instruction that varies according to the
instructional needs of students in a multi-level or multi-ability classroom.
English Language Learner: A student who is learning English as a second
language.
Focus on Form: Long & Robinson (1998) describes this as a type of language
instruction that is holistic, focused on meaning, with some grammar instruction
within the context of a focus on meaning.
Focused Task: Ellis (2003) describes it as type of second language instruction
that is designed by the teacher to create opportunities for instruction to focus on a
specific grammar topic.
Form Focused Instruction (FFI): An implicit form of grammar instruction,
within the umbrella of communicative language teaching that is associated with task
based instruction and may incorporate focused tasks (Skehan, 1998).
Independent Work Time: Bereiter et al. (2000) describes it as a 20-40 minute
period of instruction within OCR when most of the students are engaged in
independent activities, giving the teacher the opportunity to provide differentiated
8
instruction by working one-on-one or with flexible small groups of students to pre-
teach or re-teach OCR lessons.
Information-Gap Task: Ellis (2003) describes it as a a type of task based
instruction with two participants, one who holds the information and the other who
does not. To complete the task this information must be exchanged.
Jigsaw Task: Ellis (2003) describes it as a a type of two-way information-gap
task (task based instruction). The two participants each hold information needed to
complete the task, and a back and forth flow of that information is needed to
complete the task.
Minority Language Student: within the United States, one who is learning
English as a second language; this term is similar to English language learner except
that it emphasizes that the student’s primary language is not that of the dominant
culture.
Task Based Instruction (TBI): A communicative language teaching approach
whereby language learners must work together and use the second language to solve
an authentic communicative problem (Ellis, 2003).
Zone of Proximal Development: Represents skills and concepts maturating
within a learner, which may be used to solve problems under conditions of assisted
performance; Vygotsky (1986) posited that two students may have the same level of
independent performance yet have very different instructional needs because one
may have a larger zone of proximal development.
9
Limitations
Second language acquisition is a complex process that typically spans many
years, but the study took place over a period of only one month; therefore it was
difficult to extrapolate second language acquisition from such a short time frame.
Furthermore, while the purpose was to simulate the use of TBwI within the regular
school day by the regular teacher, I was unable to obtain permission from the school
principal. Since I was not the students’ regular teacher, it took place outside of the
instructional day. Therefore, I had limited ability to integrate the intervention into the
regular instructional program.
10
CHAPTER 2, LITERATURE REVIEW
Introduction
About two years ago I had the opportunity to enroll in a masters of education
program. I had been teaching for about six years and felt there to be a disconnect
between the language arts curriculum (designed for native English speakers) and the
needs of my English language learners (the term English language learner and
minority language student will be used interchangeably in this paper). I had many
questions as I began my course of study. The most important one formed the nucleus
of this thesis: how can I translate what I learn into second language acquisition for
my students?
The first part of this literature review examines the current condition of
California’s public education for minority language students (those whose mother
tongue is not English) and seeks to identify areas of imbalance. As shall be
elucidated below, it is not hollow political rhetoric to speak of California’s
educational crisis. For some it is experienced second-hand through television news
bytes. For many teachers and the communities they serve, however, education crisis
is a visceral reality.
The second part of this literature review looks at the knowledge base of
constructivist pedagogy and second language acquisition to see if there are concepts
and approaches that can be transplanted into California’s elementary classrooms of
minority language students.
11
California’s Education Crisis
There is a growing consensus that the achievement gap between students of
high and low socioeconomic status is indicative of an education crisis in California.
Barton (2005) contended that the State’s official high school graduation rate of 87
percent is really about 71 percent. Other researchers looking specifically at students
in the Los Angeles Unified School District (LAUSD) have found that only 20 percent
of English language learner students will enter high school and successfully complete
coursework that will qualify them for admission to the University of California
system. The LAUSD’s Hispanic students have a high school graduation rate of about
39 percent, one of the state’s lowest (Losen & Wald, 2005). California is second to
last among the states--above Mississippi--in terms of high school seniors who enroll
in four-year colleges (Rogers, Terriquez, Valladares, & Oakes, 2006).
The mayor of Los Angeles has characterized the problem as “the new civil
rights issue of our time” (Landsberg, 2006).
In 1996 State of California sought research-based solutions to address the
growing concerns in public education (Moustafa & Land, 2001). Researchers
presented data that OCR was effective in bringing students of low socioeconomic
status to grade level (Foorman, Francis, Fletcher, Mehta, & Schatschneider, 1998).
Ten years later, OCR had become the program of choice in California schools of
low-socioeconomic status minority language students; with only two primary
language arts programs to chose from (OCR or Houghton Mifflin Reading), about 85
percent of California school districts had chosen OCR (Manzo, 2004).
12
It has been called a civil rights issue because the dropout rate has
disproportionately affected certain minorities. This problem is not restricted to the
Hispanic students of the LAUSD but has affected minority language students across
the state. This is a significant problem because about 24 percent of students in the
public schools are learning English as a second language; about 1.5 million minority
language students were enrolled in California public schools (out of a total
enrollment of 6.3 million), with especially high concentrations in certain schools
(Rumberger, Gándara, & Merino, 2006).
The success in achieving English language proficiency varies according to
many factors. For example, the California Legislative Analyst’s Office found that
approximately 50 percent of students whose first language is Spanish redesignate as
English proficient after 6.7 years of instruction versus 3.6 years for students whose
first language is Mandarin (Warren, 2004). These results are also consistent
Cummins’ (1994) timeframe of 5-7 years to achieve cognitive academic language
proficiency.
English language learners, especially inner-city Hispanic ones, have higher
high school drop out rates yet those who rapidly acquire English language
proficiency do not. Thus, one response to this problem has been to advocate change
in education policy, especially at the elementary school level where students begin
their journey toward English language proficiency.
13
Changes in Education Policy
The adoption and implementation of OCR has taken place against a backdrop
of two radical shifts in teaching methods within California’s elementary public
schools (kindergarten through grades 5/6). The first has been the decline of bilingual
education. As of January 1999, California’s Proposition 227 severely limited public
schools’ ability to offer bilingual education programs. At the national level, the 2002
repeal of the Federal Bilingual Education Act reinforced Proposition 227’s intent of
all-English instruction (Crawford, 2004). Academia within second language
acquisition was among the strongest proponents of bilingual education (Crawford,
2004; Krashen & Biber, 1988; Thomas & Collier, 2002; Cummins, 1994). With the
rise to prominence of all-English instruction, the theoretical and research base of
second language acquisition has largely disappeared from the discourse of K-12
public education in California. The primary exception to this has been specially
designed academic instruction in English, but these strategies are designed to support
content instruction, not English language acquisition (Genzuk, 2003).
Later, I shall review the literature regarding OCR along with select topics
from second language acquisition and constructivist pedagogy. However, it is
important to bear in mind the broader historical context of changes in education
policy. It has been these changes that have led to a mismatch between needs of the
curriculum and those of minority language students.
The second radical shift has been characterized as the reading wars, a
polarizing struggle between whole language and phonics-based approaches (Pearson,
14
2004). Whole language has been difficult to define, however it tends to emphasize a
holistic approach based upon constructivist principles (Richards, Rodgers, & Swan,
2001). In contrast, a phonics-based approach posits that many students need explicit
instruction in the sound-spelling system of English in order to attain literacy. Aside
from phonics, there were four casualties in the wake of widespread use of whole
language: skills instruction, strategy instruction, text structure instruction, and
content area reading; in addition, there was very strong resistance to standardized
accountability measures (Pearson, 2004).
As with bilingual education, the triumph of phonics over whole language
took place first not in classrooms but in the halls of power. In 1996 the California
Board of Education placed OCR on the state textbook adoption list over objections
voiced by teachers on the Instructional Resources Evaluation Panel (Moustafa &
Land, 2001). Injecting explicit instruction of phonics, skills instruction, strategy
instruction, text structure instruction, content-area reading, and accountability
measures into the curriculum may have been needed to restore a balance to literacy
instruction. Unfortunately, with the triumph of a phonics-based approach (e.g., OCR)
and direct instruction pedagogy (see below), there came a paradigm shift, “... and it
was not clear whether there was a place for constructivist pedagogy in general or
whole language in particular in these new conversations.” (Pearson, 2004, p. 228).
The catalyst of this investigation was a feeling. The puzzle of literacy
instruction in my classroom seemed to be missing a piece. I had been told at
LAUSD-sponsored professional developments that whole language was bad and
15
explicit phonics instruction was good. Using OCR, I was told, would lead to faster
decoding speed, which was equated with English literacy success. Indeed, using
OCR my English language learners did receive grammar instruction, strategy
instruction, explicit phonics, as well as exposure to a variety of genres and text
structures. Yet those same English language learners said to me, “I don’t understand
the story.” and “I don’t like writing because nobody reads it.” I took note of the fact
that OCR was primarily designed for native English speakers. Later, a review of the
literature led me to suspect the missing piece to be a holistic approach that would be
meaning centered, based upon constructivist pedagogy and principles of second
language acquisition.
The Research on Open Court Reading
The literature on OCR supports the notion that it is a balanced and research-
based literacy program that emphasizes explicit instruction of phonics and phonemic
awareness. However, it also points to what is missing piece of the literacy puzzle for
English language learners: a holistic approach based upon constructivist pedagogy.
The first major study on OCR took place in Texas schools in the early 1990s.
The researchers’ hypotheses was that explicit (as opposed to implicit) phonics
instruction would result in greater phonemic awareness for primary students at risk
for reading failure, and that this growth would influence academic growth: “The
results of this research clearly indicate that early instructional intervention makes a
difference for the development and outcomes of reading skills in the first- and
second-grade children at risk for reading failure.” (Foorman et al. 1998, p. 51). The
16
goal of the study was to compare the effectiveness of three different approaches:
direct code, which was explicit instruction of phonics and phonemic awareness; (2)
implicit code, a literature-based curriculum with implicit phonics instruction; and (3)
embedded code, an intermediate approach. The study gauged success of literacy
instruction in terms of reading fluency, defined as the number of correct words (total
words read minus errors) the student could sound out from a passage not previously
encountered. The researchers found that students with low initial levels of phonemic
awareness experienced minimal decoding growth with implicit phonics instruction in
contrast to significant growth with explicit phonics instruction. Implicit phonics
instruction is associated with the whole language approach.
There has been much debate regarding the reliability and validity of the
study. It took place at three different school sites. The site using implicit code also
had the highest percentage of students on a free lunch program (i.e., they had the
lowest socioeconomic status). As noted above, a pre-publication version of this
report was presented to the California State Assembly Education Committee and was
a major force behind the adoption of OCR.
The National Reading Panel (2000) confirmed these findings. The report was
a meta-analysis of the existing research. As such, it only included quantitative studies
that had an experimental or quasi-experimental design (experimental group, control
group, statistical significance, etc.). Qualitative research studies were excluded. The
report had many recommendations. Primary among them was that explicit instruction
17
of phonics and phonemic awareness in the early grades were important elements of a
balanced literacy program.
Maddahian (2002) used quantitative methods to examine the efficacy of OCR
and found it to be effective in raising test scores for low-performing second grade
students. The study compared OCR with Success for All, another commercial
program that also has a strong emphasis on explicit instruction in phonics and
phonemic awareness.
Izumi, Coburn, & Cox (2002) looked at model schools to see how their
success could be replicated. They examined eight schools with high academic
achievement and low socioeconomic status. The researchers observed classrooms
and interviewed principals about what factors had contributed to their success. Direct
instruction was found to be a key element to school improvement. “An important
advantage of using a scripted curriculum such as Open Court is that all teachers
adhere to a pacing schedule that requires that lessons are set according to a strict
timetable. Thus, for instance, on a given day every second-grade teacher will be
teaching the same lesson.” (Izumi et al., p. 13).
This study is significant in that it highlights one potential weakness of OCR:
lack of differentiated instruction. Children who have differing levels of language
development have differing needs. A fourth grade child recently arrived from
Mexico City and at grade level has differing needs from a student who previously
lived on a Mexican hacienda and arrives with no prior formal schooling experiences.
Both of them have very different needs from an African-American child who arrives
18
to school speaking a non-standard English dialect and the second generation
Hispanic child whose parents only speak Spanish.
Williams et al. (2005) surveyed principals and teachers at 257 schools, with
normal distribution of low-, mid-, and high performing schools. One of their
conclusions was the importance of a coherent standards-based instructional program.
They noted a correlation between high performing schools and districts that played a
strong role in curriculum development. “These principals report that the district has
clear expectations for student performance aligned with the district’s adopted
curriculum and that it evaluates the principal based on the extent to which instruction
in the school aligns with the curriculum.” (Williams et al., p. 18). This is significant
because it shows the pressure placed upon school administrators to emphasize
uniformity rather than differentiation.
Direct Instruction
As noted above, one of the major policy recommendations of the National
Reading Panel (2000) was the importance of explicit instruction in phonics and
phonemic awareness for the early grades. Much of the literature on OCR is infused
with direct instruction pedagogy. Sig Englemann, the famed direct instruction
pioneer at the University of Oregon, characterized direct instruction as having a
scripted and rapid-paced instruction (Izumi et al., 2002). Englemann authored SRA’s
Reading Mastery, a highly scripted phonics based program which was later expanded
into OCR. O'Neill (1988) reviewed the literature on teaching effectiveness and found
direct instruction to be “... highly associated with increased gains among primary
19
children from working and middle class backgrounds is a common, almost universal
conclusion of recent research. However, the effects appear to be more pronounced
for low-ability pupils with an external locus of control.” (O'Neill, p. 173).
In a more recent study, Conner, Morrison, & Katch (2004) looked at the
differential effects of implicit versus explicit instruction as well as teacher-centered
versus child-centered instruction. They analyzed the effectiveness of these different
types according to students’ fall and spring decoding and vocabulary abilities. They
concluded that students with low initial vocabulary and low decoding skills benefited
from teacher-managed explicit phonics instruction (TME) followed by progressively
more implicit instruction during the school year. Students with low initial decoding
but high vocabulary benefited most from TME, child-managed implicit instruction,
and lots of opportunities to do independent reading/writing activities. This study
highlights the importance of flexibility within the classroom to respond to the diverse
needs of the students. Clearly, some studies have established statistically significant
advantages to direct instruction. At the same time, not all students respond equally
well to the same methodology.
The research base critical of OCR is both quantitative and qualitative, having
both academic and sociopolitical implications. Literature critical of OCR in terms of
academics have focused on the complex nature of literacy as well as the need for a
meaning-centered approach (Moustafa & Land, 2001; Peck & Serrano, 2002;
Wilson, Martens, Arya, & Altwerger, 2004). In terms of a sociopolitical dimension,
Achinstein & Ogawa (2006) as well as Achinstein, Ogawa, & Speiglman (2004)
20
have identified the emergence of a two-tier educational system in California: (1) a
constructivist and student-centered curriculum that stresses academic freedom,
creativity, and high-level thinking skills for schools in higher socioeconomic areas;
and (2) a direct instruction textbook-centered curriculum that stresses rigid
uniformity of instruction, narrow measures of literacy, and lower-level thinking
skills. Especially disturbing is the implication that children of low socioeconomic
status (within California urban schools, often minority language students) have a
fundamentally different learning style that requires a regimented and structured
pedagogy emphasizing lower-level thinking skills.
Wilson et al. (2004) found that implicit phonics instruction embedded within
meaning-centered instruction results provides superior results over explicit phonics
instruction because students develop better reading comprehension. They studied 84
urban students of low socioeconomic status using three programs: (1) Direct
Instruction, a predecessor to OCR; (2) OCR; and (3) Guided Reading, a literature-
based program adapted from Irene Fountas and Gay Sue Pinnell. The students were
tested using the phonics subtest of the Woodcock-Johnson Psycho-Educational
Battery. They also conducted observations and teacher interviews. They found that
the percentages of miscues at the Guided Reading site was lower because there was a
greater concern for meaning. It should be noted, however, that no English language
learners were included in the study.
Moustafa & Land (2001) found no significant advantage to using OCR with
economically disadvantaged students. In contrast to previous studies, they looked at a
21
more comprehensive measure of literacy: second through fifth grade scores on the
SAT/9 (a California achievement test for elementary students). From the LAUSD,
159 schools were selected and categorized according to percentage of students on the
free lunch program and which English language arts program they were using (Non-
OCR, OCR, and long-term OCR). For second grade students, they found no
statistically significant difference between the three programs. However, when
looking at all students (grades 2-5), both types of OCR schools were significantly
more likely to be in the bottom quartile of SAT/9 academic achievement.
In the same study, they looked at teacher ability to differentiate instruction
and respond to the individual needs of the learners. The OCR teacher’s manual
contains a plethora of activities and does not appear to be a scripted or rigid program.
However, in every district-wide adoption we are aware of in California, the
state in which our study took place, teachers are required to complete every
activity described in the teachers’ manual with the entire class, whether it is
appropriate or not, and to do it at a prescribed pace (i.e., so many lessons
within so many days), whether it is appropriate or not. (Moustafa & Land,
2001, p. 8).
This is significant because the validity of the research supportive of OCR
assumes a program that allows the teacher to respond and adapt to the needs of the
students. Furthermore, one of the objections encountered in the literature is not to
OCR per se, but to a rigid and inflexible implementation that does not allow for
differentiation of instruction.
Peck & Serrano (2002) studied approximately 100 teachers in a university
program and identified concerns regarding differentiating instruction for English
22
language learner students (ELLS). This qualitative study was based on their records
working with both new and experienced teachers (some were in a teaching credential
program and others were returning for their master degrees). They looked at lesson
observations, field notes, and surveys. Much of their paper discussed broad themes
that had been identified as concerns by the participants and through observations:
On the positive side, the network of support for ELLS can be strong and
effective, depending on the teacher and coach. The curriculum definitely
provides clear guidelines and rules. On the negative side, we heard numerous
examples of children who did not understand the teacher’s oral language or
the stories they were asked to read. Rather than being asked to accommodate
to the children’s levels, teachers were asked to follow a script, treating all
students the same. (Peck & Serrano, p. 6).
More recently, Achinstein & Ogawa (2006) highlighted the growing pressure
teachers face when attempting to differentiate their instruction according to the needs
of their students:
In our two case studies, district and school administrators enforced fidelity to
improve the academic performance of students by requiring teachers to
adhere to Open Court’s pedagogical script, use only materials provided by the
program, and cover the material at the prescribed pace. Fidelity left little or
no room for teacher discretion and thus suppressed teachers’ reflection and
discussion. (Achinstein & Ogawa, p. 54).
Achinstein et al., (2004) contrasted teacher enculturation in two different
school districts. They concluded that one school district, faced with limited resources
and large numbers of minority language students, adopted a culture that discouraged
higher-level thinking skills both for students and teachers. It was a behaviorist
pedagogy that was sold as teacher-centered but was textbook-centered in actuality.
The other school district, with abundant resources, embraced a constructivist
23
pedagogy, focused on student-centered instruction, and cultivated higher-level
thinking skills at all levels, from the student right up to the superintendent.
For Schroeder (2006), "Our charge [as public school educators] is to
fundamentally improve the interaction between the teacher and the student to create
critical thinkers prepared to participate in a diverse and complex society." The next
section examines a pedagogy that addresses this need for quality of interaction
between teacher and student.
Constructivist Pedagogy
Vygotsky (1986) developed the concept of ZPD for teachers to assist the
child in developing a mature schema of scientific concepts and logical thinking,
which he refers to as a fabric of concepts.
The ZPD represents skills and concepts a student can successfully access
under conditions of assisted performance. It is an important theoretical basis and
rationale for differentiated instruction. Every classroom has students working at
different ability levels, not only in terms of what they can do independently but also
what they can do under conditions of assisted performance.
The concept of ZPD highlights the need for formative assessments that are
forward looking (measures of assisted performance) rather than backward-looking
(measures of skills/concepts previously mastered), as can be seen from Vygotsky’s
concern about the dubious nature of intelligence tests. Two students who might test
at the same level on an assessment of previously mastered skills might have very
different instructional needs.
24
Let us take a simple example. Suppose I investigate two children upon
entrance into school, both of whom are ten years old chronologically and
eight years old in terms of mental development ... These children seem to be
capable of handling problems up to an eight-year-old’s level, but not beyond
that. Suppose that I show them various ways of dealing with the problem ...
Under these circumstances it turns out that the first child can deal with
problems up to a twelve-year-old’s level, the second up to a nine-year-old’s.
Now, are these children mentally the same? (Vygotsky, 1978, pp. 85-86).
Of course, the teacher-practitioner would ask a slightly different question: are
these children’s instructional needs the same? As with teacher-practitioners,
Vygotsky was concerned not only with the level of instruction but also the quality of
interaction between the teacher and student. Imitation can have a negative
connotation (cf. accommodation), but Vygotsky saw it as an effective vehicle for
assisted performance provided that it correlated with the child’s the ZPD:
But recently psychologists have shown that a person can imitate only that
which is within her developmental level. For example, if a child is having
difficulty with a problem in arithmetic and the teacher solves it on the
blackboard, the child may grasp the solution in an instant. But if the teacher
were to solve a problem in higher mathematics, the child would not be able to
understand the solution no matter how many times she imitated it. (Vygotsky,
1978, p. 88).
Let us extrapolate this concept to the area of second language acquisition.
Each language learner has uniquely different combinations of emergent grammatical
structures and vocabulary. To push students to ever higher levels of independent
performance, the teacher must choose a topic and method of interaction with the
student. Each student may have a unique ZPD. Interaction below this zone would
likely result in no new learning. Interaction above this zone would likely have the
25
result that the student neither learns nor understands, no matter how many times the
teacher is imitated.
Again, Vygotsky would argue that ZPD should be used as a tool to orient
teachers towards those functions of the student “currently in an embryonic state”
(Vygotsky, 1978, p. 86) rather than ones already mastered. Teachers should have
high expectations and focus on what students are in the process of learning rather
than what they have already mastered. However, to go beyond the ZPD could result
in mechanical imitation and absence of learning.
In some cases, the concept of ZPD has lost some of its force in classrooms as
instruction becomes increasingly passive. One way of looking at this process is
commodification, wherein schools begin to be structured within an assembly-line
construct toward the production of a commodity: school knowledge. “Teachers and
schools, often coerced by governments and education authorities, have, over many
years, evolved a school curriculum based on the production of endless routine
exemplar problems which facilitate short-term reproduction for assessment
purposes.” (Beveridge, 1997, p. 29). Explicit and regimented whole-group phonics
instruction may have a valid place in the school day, but Vygotsky would probably
say that the most effective instruction stems from assisted performance that is
differentiated to the minority language student’s given ZPD for second language
acquisition. This would be instruction centered upon student needs rather than the
dictates of a pacing plan and textbook designed for native English speakers.
26
According to Piaget’s constructivist pedagogy, the student is presented with
several alternatives when confronted with new information or ideas that are not
consistent with what has already been internalized. This feeling of cognitive failure
can be repressed or laughed off; alternatively, it can be acknowledged and a process
of cognitive restructuring (assimilation) begins.
The pressure to perform in certain school situations, however, might well
place a premium on students' accommodations to teachers' descriptions of
correct procedures or answers. In many cases, these accommodations attain
short-lived success and do not ensure a sufficient level of understanding to
guarantee that the successful performance or correct answers will be
replicated at subsequent points in time. (De Lisi & Golbeck, 1999, p. 9).
This reiterates the point made by Vygotsky that certain types of learning may
result in short-term gains if a student is working above the level of meaningful
assisted performance or there is pressure to simply repeat the correct answer. In
contrast, interaction that is meaning-based and student-centered may create an
opportunity to acknowledge the cognitive failure from a breakdown in
communication between the second language learner and teacher.
To conclude, Vygotsky’s ZPD presents us with an apprentice model of
learning. Through various forms of assisted performance, the student’s emerging
ability may maturate until it can become internalized. Crucial to this process would
be two factors. First, the teacher would accurately evaluate the student’s ZPD;
although assessments that measure independent performance are certainly valuable,
the best measure of ZPD will be actual real-time student performance under
conditions of assisted performance (i.e., teacher observation and reflection). Second,
27
the teacher should have the discretion to act upon this information to deliver
differentiated instruction that is truly targeted to the student’s ZPD; at least some
portion of the day should be guided neither by scripts nor curriculum pacing plans
but instead by the true individualized needs of the student. Meanwhile, Piaget’s
constructivist model of learning suggests that meaningful long-term learning does
not come from rote learning but from meaningful opportunities to experience a
breakdown in communication, acknowledge the cognitive failure, and then begin the
process of restructuring the emerging English competency.
The constructivist pedagogy of Piaget and Vygotsky examines the process of
knowledge and concepts acquisition in general as well as the type of teacher-student
interaction that result in student learning. It also serves as a foundation for the
subsequent development of current second language acquisition theory and practice.
The following section deals with second language acquisition and how teacher-
student interaction can meet the needs of second language learners.
Second Language Acquisition
Much investigation has taken place to understand the differences and
similarities of first and second language acquisition. Krashen (1982) made a
distinction between language learning and acquisition, stating that traditional
grammar-based approaches often lead to knowledge of the language as an object of
formal study rather than communicative competence. In contrast, acquisition of the
second language is a process similar to what children experience with their first
language. The logical conclusion of such a premise is that since children do not
28
receive grammar lessons from their mothers/caretakers, neither should older
language learners interested in acquiring a second language. A teacher using Krashen
& Terrell’s (2000) natural approach would provide lots of comprehensible input in a
relaxed learning environment during the initial stages of language acquisition.
However, both Long (1996) and Swain (1985) were concerned that input
alone would not lead to mastery of the language, especially in terms of expressive
language and grammatical competence. Thus, continued a central debate within
academia of second language acquisition: What role does grammar instruction have
within the classroom? What is the relationship between receptive and expressive
language development? None of these questions have been conclusively answered. In
part, this is due to the high failure rate in adult second language learning when
contrasted with almost universal success of first language acquisition (Bley-Vroman,
1988).
Task-Based Instruction
One proposed answer to the questions raised in the above paragraph is TBI
that includes FFI. It represents a middle ground between naturalistic forms of
communicative language teaching focused on comprehensible input and traditional
language teaching focused almost exclusively on grammar. According to Ellis
(2003), although the emphasis of TBI is on oral communicative competence, it can
encompass all four domains of language (listening, speaking, reading, and writing);
while grammar exercises (associated with traditional language teaching) prompt
students to learn target language forms, and tasks guide students to actually use the
29
target language. Skehan (1998) noted that tasks emphasize primacy of meaning and
present a communication problem that needs to be solved.
Probably the most fundamental mind-shift of TBI is the concept of linking
classroom instruction with the real-world importance of language: communication.
Smith (1988) observed that:
Anything a child is not interested in doing should be modified or avoided.
Forcing a child into boring or painful activity will merely teach the child that
the activity is boring or painful, no matter how good we think it is for the
child. Anything with a mark attached should be avoided. Children quickly
learn that many school activities are worth doing only for the grade, and
when they learn that, they learn that the activity is intrinsically worthless.
(Smith, p. 15)
For Lightbown (1998), the segregation between language instruction and
language use can become a self-fulfilling prophesy when language knowledge and
actual use are segregated in a way that language learning does not lead to
grammatical competence under conditions of actual communication.
Form-Focused Instruction
The concept of FFI evolved in conjunction with TBI as a method of
developing grammatical competence within the context of a communicative language
approach. FFI is not one single approach, but, instead, is a continuum of approaches
ranging from explicit and planned to implicit and reactive. Long & Robinson (1998)
defined FFI as an occasional shift of attention from meaning to grammar, similar to
what happens in real life when there is some sort of breakdown in communication.
For example, Lightbown (1998) refers to the concept of negotiation of form in terms
30
of a French immersion teacher who is very skilled in getting learners to correct their
own oral grammar mistakes through incisive questions.
Krashen (1982) developed the monitor hypothesis to account that the
awareness of the rules of a language could allow a person to correct mistakes yet not
lead to fluent native-like production. Skehan (1998) built upon it by positing that the
brain has a dual coding system: (1) a rule-based and an (2) exemplar-based system
for storing language information. While the rule-based system is flexible and logical,
it does not lead to fluent production. Fluency in speech comes from the exemplar-
based system. Unlike Krashen, Skehan posited that when the language learner is
pushed to produce output by means of his rule-based system, what s/he speaks in turn
becomes input for the exemplar-based system. Thus, over time awareness of the rules
of language can translate to fluent and grammatically correct language proficiency.
This stands in contrast to Krashen’s position that language learning does not lead to
language acquisition.
In a similar vein, Long’s (1996) interaction hypothesis developed the
importance of selective attention, based to a large extent upon the noticing
hypothesis of Schmidt (2001). The interaction hypothesis states that a second
language learner does not attend to all features of the target language input, but in the
context of a conversation between a native speaker and language learner, the
feedback received can prompt the language learner to notice certain features that
might otherwise have gone unnoticed. Thus, TBI/FFI represents teacher-planned
interactions and discussions in class that lead to language learning; as a result, the
31
students are led to notice their grammatical errors that cause communication
problems. It is this noticing that leads to awareness and subsequent language
proficiency.
Multitudinous studies of TBI/FFI have examined whether prompts or recasts
are more effective (Ammar & Spada, 2006; Doughty, 2001; Ellis, 2006b; Harley,
1998; Lyster, 2004). Recasts are generally defined as more subtle means of
correction because the native speaker repeats what the language learner said, but
without grammatical errors. In contrast, prompts use a variety of techniques to more
explicitly draw attention to the error and prompt the language learner to restate and
self-correct. Other studies have looked at the efficacy of TBI more generally
(Mackey, 2006; Pica, 2005; Swan, 2005).
Harley (1998) examined the use of FFI with second grade language learners.
She used high-interest tasks to raise consciousness in the correct use of a specific
grammar form. She concluded that student attention was selective and limited. High-
interest and meaningful activities tended to generate the best results. Furthermore,
when students were given activities involving a large amount of new vocabulary,
they tended to focus on grammatical forms much less.
Doughty & Varela (1998) examined the use of FFI with recasts in English as
a second language science and math classes, grades 6-8. Within the context of
students reporting on their results to the class, there were many opportunities for the
teacher to use recasts to guide the students into correct usage of the past tense. The
study indicated that recasts were effective in improving use of the past tense both in
32
writing and speaking. However, there was some question as to whether these results
would result in long-term language gains.
Doughty (2001) conducted a meta-analysis of studies comparing prompts and
recasts. She concluded that recasts tend to be used for a variety of purposes and
forms in language immersion classrooms, and that this ambiguity weakens the
effectiveness to target grammar usage. In contrast to abandoning recasts in favor of
prompts, she recommended a reexamination of how recasts are used in the class
since this type of subtle error correction can be a valuable element in guiding the
student to improved language proficiency.
Lyster (2004) looked at the use of prompts and recasts with fifth grade
language learners. He looked at FFI with prompts, with recasts, and with no
feedback. His results indicated that FFI was more effective when combined with
prompts in learning the targeted grammar form (noun gender).
Ammar & Spada (2006) tested the use of prompts and recasts with sixth
grade language learners, targeting use of the possessive determiners (his and her) in
written and oral tasks. They found that high proficiency language learners benefited
equally from both while low-proficiency language learners benefited more from FFI
with prompts. The high/low proficiency categories were based upon pretests of the
targeted grammar form.
Ellis (2006b) conducted a survey of the research on recasts. He found that
recasts tend to be used in a variety of ways by teachers and researchers, ranging from
very implicit to very explicit. They can be used in a way to provide positive evidence
33
(correct examples of language use) or negative evidence (drawing attention to the
grammatical mistake). Further, he found that FFI with prompts was most effective in
improving grammar during writing tasks.
Mackey (2006) investigated the connection between FFI and noticing, and the
extent to which noticing can improve language acquisition. She found that there was
a correlation between student reports of noticing the forms and second language
development, although this correlation was stronger for certain forms (questions and
plural forms) than others (past tense). However, Mackey cautioned that because of a
small sample size and complexity of what is being measured, it would be difficult to
make definitive conclusions.
Clearly, a lot of research has been conducted on FFI/TBI, however much of it
has been experimental and theoretical in nature. Pica (2005) analyzed the research
conducted to date and examined ways in which FFI and TBI could be combined
through information-gap tasks in the classroom. She asked how these tasks could
assist second language learning, retain classroom authenticity, and adhere to the high
standards of research. Another important concern was to be able to look at longer
term applications of FFI and TBI in the classroom. She concluded that the
information-gap task was very useful as an authentic class activity for teachers and as
a research tool for academia.
Swan (2005) took a more critical look at TBI. He raised valid methodological
concerns regarding the studies that support the use of TBI in the classroom.
Furthermore, he observed that there was no research to support the contention that
34
traditional grammar-based approaches had failed, as had been alleged by supporters
of TBI. He concluded that although TBI should not be used as an exclusive guide for
constructing the syllabus for language learners, it certainly could be another resource
used by language teachers to respond to the diverse needs of the language learners.
As I examined this literature, two points became increasingly clear: (1) more
research was needed to understand how TBI/FFI could be used as an effective tool
by a language teacher such as myself; and (2) a formulaic approach was not
warranted.
Task-Based Writing Instruction
One line of inquiry that had not been previously examined within the research
was the use of the writer’s conference as a jigsaw task, a two-way information-gap
task. I have coined the term task-based writing instruction (TBwI) because, as shall
be elucidated below, the literature of second language acquisition, applied linguistics,
and English language arts (literacy instruction to native-English speakers) makes
reference neither in name nor substance to this transformation of a stage of the
writing process into a tool for second language acquisition.
As mentioned previously, Lightbown (1998) saw one manifestation of TBI/
FFI as a teacher-student dialogue whereby the student would self-correct oral
grammar mistakes through incisive teacher questions. A one-on-one conference
between the teacher (the native or native-like English speaker) and student (the
English language learner) to move a rough draft through the stages of the writing
process could be vehicle for incorporating TBI/FFI into the classroom.
35
I embarked upon a thorough review of the literature to search for the
application of TBI/FFI within the context of the writing conference, or perhaps
something akin to this without the use of these terms. I consulted several
academicians in the area of second language acquisition and applied linguistics. I
conducted multiple searches of the ERIC database using various iterations and
combinations of terms such as writing, conference, task-based, form-focused, TBI,
FFI. I consulted the references section of every single book and journal article cited
in this paper. I found nothing.
Regarding the writing process, the literature of second language acquisition
and applied linguistics essentially parallels that of literacy instruction for native
English speakers. For example, Ferris’ (2002) recent treatment on the subject
includes both her own research and an exhaustive review of the literature. Teacher
involvement in editing student work is treated as a temporary support that should be
phased out as the second language learner is taught to independently self-edit and
self-correct writing pieces. She addresses the questions of what to correct, what not
to correct, and how to get students to notice specific categories of errors through
systematic instruction of error classification and various editing strategies (e.g.,
having the students make multiple passes of editing their work, each time looking for
specific error types). This strategy bears a similarity to one manifestation of FFI:
input flood. With this type of FFI, the student is given a written language sample that
has multiple instances of a particular language form (often made conspicuous
through a technique such as highlighting or underlining) so that noticing--and by
36
extension, acquisition--of the grammatical form will occur. However, the writing
conference itself is not treated as an authentic language exchange between a native
speaker (teacher) and non-speaker (student).
The writer’s conference represents one of the few authentic opportunities for
a one-to-one interaction between the language learner student and the teacher of a
large mixed-ability classroom. The writer’s conference is a jigsaw task (a two-way
information-gap task) with true negotiation of meaning between a native and non-
native speaker centered around a written work. The second language learner enters
into a dialogue with a native speaker in order to overcome the limitations of the
interlanguage (i.e., student emergent abilities in the second language) encoded within
the paper. Furthermore, it is possible for the teacher to construct the writing prompt
in a manner that requires or strongly encourages the use of a particular linguistic
form. In addition, if instead of marking up the student’s paper with proofreading
marks, the teacher produces a final draft of the student’s paper as a product of the
writing conference, then an additional opportunity is created for the language learner
student to notice differences between his/her interlanguage and standard English.
This would be accomplished by requiring the student to compare the final draft (a
product of negotiation of meaning during the writing conference) and the original
draft s/he brought to the writing conference; this is accomplished through a followup
activity that requires the student to mark up the original draft with proofreading
marks by comparing it to the final draft created by the teacher.
37
Conclusion
In conclusion, a large body of research has highlighted the achievement gap
between minority language and native English students within the California public
education system. Different groups of minority language students experience
different levels of success in achieving English language proficiency. Minority
language students experience significantly higher levels of school failure than native
English students.
One response to this problem has been a rigid implementation of OCR, based
upon research that demonstrates the importance of early phonics and phonemic
awareness instruction. Some have contended this reading program (designed for
native English students), or perhaps the manner in which it is implemented, does not
fully address the diverse needs of minority language students.
OCR is associated with the reemergence of a direct instruction pedagogy that
is behaviorist in origin. An examination of learning principles of Piaget and
especially Vygotsky leads to the conclusion that one way to meet the diverse needs
of minority language students would be to provide targeted differentiated instruction
for at least some part of the day. OCR already has built within it a 20 to 40 minute
block of independent work time when most students are given the opportunity to
work independently while the teacher works with flexible groupings of students to
provide differentiated instruction; however, this differentiated instruction typically
consists of pre-teaching or re-teaching OCR lessons and pedagogy.
38
An alternative would be to draw upon principles and practices from the field
of second language acquisition to create truly differentiated instruction, consistent
with the constructivist pedagogy of Piaget and Vygotsky. One methodology from
second language acquisition research and theory is TBI/FFI. A significant line of
inquiry within FFI/TBI has been the information-gap task. Information-gap tasks are
constructed so that the second language learner needs to interact with someone else
(preferably a native speaker) in a manner that simulates an authentic real-world use
of language as a tool for communication and exchange of information. A task is TBI/
FFI when it includes occasional switches towards focus on form (i.e., focus on
grammar) that maintains the primacy of meaning.
While much of the theory and research of TBI/FFI has been positive, it has
also been mixed in terms of understanding how to translate this research base into
teaching practice that is an vehicle for second language acquisition in the classroom.
This review of the literature points to the need for additional research into the use of
TBI/FFI as a resource for language teachers to respond to the diverse needs of the
students in their charge.
A heretofore unexplored avenue is TBwI, the transformation of the teacher-
student writing conference into a jigsaw task consistent with the principles of TBI/
FFI. The next chapter outlines the plan and methods formulated to answer my
research questions, which crystalized during my review of the literature.
39
CHAPTER 3, METHOD
Introduction
As the outlines of what would become this study and thesis paper began to
form, I realized that a purely quantitative methodology would be inadequate for the
questions I wanted to pose. Traditional measures of student achievement examine
what the language learner is able to do independently. They measure the fruits of the
interaction between student and teacher (i.e., second language acquisition under
conditions of independent performance), but not the interaction itself. This is not
problematic when the learning cycle is short. It is not problematic when a single skill
is taught and assessed in isolation. However, language acquisition is far more
complex and challenging. It is a system of many interrelated and interdependent sub-
skills that are generally acquired over the course of many years. It is a holistic
process that takes place every time the student is posed with a language task, whether
in a research facility, public school, or train station. Swan (2005) rightly pointed out
that the literature has not demonstrated that TBI/FFI, neither in a clinical nor
classroom setting, results in meaningful long-term second language acquisition.
Indeed, much of the research into TBI attempted to quantify short-term gains
(Ammar & Spada, 2006; Doughty, 2001; Doughty, & Varela, 1998; Ellis, 2006a;
Harley, 1998; Lyster, 2004; Mackey, 2006).
Given that language acquisition is a complex process that can span many
years (Cummins, 1992), creating a rigorous quantitative study that successfully
controls for dependent and independent variables would be both problematic and
40
unethical. I would have to control and/or monitor all instances of language
acquisition over a period of many years. This would mean exerting a level of control
over the subjects that would be contravene accepted ethical standards. It would
almost certainly also be unacceptable to the subjects and their parents. Thus, I
decided to explore other research paradigms.
Several researchers, especially in the area of applied linguistics have taken to
Vygotsky’s desire to focus on the quality of interaction and feedback between
student and teacher. This area of applied linguistics is known as discourse analysis.
Cazden (2001) noted that much of this research has been directed toward an analysis
of improving quality of interactions between teacher and student or raising
consciousness of the powerful impact of peer assisted learning.
Haneda (2004) successfully examined the interactions between student and
teacher, with a focus on understanding how the student could take a more active role
in the construction of meaning within the writing conference. The study consisted of
a transcription and analysis of three writing conferences for each of nine participants.
The study was mixed methods in that there was some quantitative analysis, but
primarily it was a qualitative analysis of the interactions between teacher and student
within the writing conference.
Gibbons (1998) also conducted a qualitative analysis of student-teacher
interactions, primarily ways in which teachers could expand upon & recast student
language during whole group discussions of science experiments so as to enhance
critical thinking and academic language proficiency. Although the method section
41
did not explicitly state the methodology used, the discussion section indicated that a
transcriptions of classroom discourse had been analyzed using purely qualitative
methods.
I found that Haneda’s (2004) blend of quantitative and qualitative data
analysis to be appropriate for the questions posited in this study. Thus, I used a
similar mixture of methodological approaches, both qualitative and quantitative.
Quantitative measures were used to determine the degree to which TBwI could be an
effective vehicle for providing differentiated instruction. Again, a major impetus for
this investigation was the degree to which TBwI could address each subject’s unique
instructional needs (i.e., be an effective vehicle for differentiated instruction) in
contrast to the efficacy of the instruction itself (i.e., the amount of measurable second
language acquisition). Qualitative data analysis was used to examine how principles
of second language acquisition and constructivist pedagogy could be used and
incorporated within OCR; in addition it was used to examine evidence for an impact
on second language acquisition.
Research Questions
The goal of the current study was to explore the impact of incorporating
principles of TBI/FFI, consistent with principles of second language acquisition and
constructivist pedagogy, into written language instruction for third grade English
language learners. The specific research questions addressed by this study were:
(1) What are the issues involved with using TBwI as communicative
language teaching consistent with principles of TBI/FFI & constructivist pedagogy?
42
(2) How does TBwI impact the degree of differentiated instruction within a
mixed ability classroom?
(3) How does TBwI impact second language acquisition?
The research questions were answered through a combination of quantitative
and qualitative data analysis from the following data sources: (1) transcripts of
writing conferences; (2) subject work samples; and (3) interviews. Whereas
traditional research has focused on using recasts and prompts of oral utterances, this
study focused on recasting written sentences, thus it was necessary to examine the
subject work samples in conjunction with the transcripts of the writing conferences.
The interviews were a secondary source of data to examine changes in the subjects’
performance.
Research Context
School Site Context
The study was conducted in a single urban public school, which has students
from kindergarten through eighth grade. The students were overwhelmingly Hispanic
(97.8 percent of about 1,000 students as of the 2005-2006 school year).
Approximately 50 percent of the students were classified as English language
learners with Spanish as the first language. Due to overcrowding, the school operated
on a multi-track calendar; therefore, at any given time approximately one-third of the
students were on vacation. This also meant that there were 163 instructional days
instead of 180. Although the school day was lengthened to compensate for this,
school district instructional pacing plans necessitated compressing 180 lessons into
43
163 days. This meant that there was even greater pressure to deliver a standardized
curriculum since teachers had fewer days to teach the same number of textbook
lessons.
There were 54 teachers on staff, a little under half of whom have over five
years teaching experience, which was consistent with my teaching experience and
educational qualifications. I selected this school site in part because I had an insider’s
perspective (see below, Researcher Participant) at this school site since I had taught
there for about five years when the study began. An emic perspective is an important
characteristic as one of the research questions asked how authentic communicative
exchanges within the context of a writing conference between teacher and student
can impact second language acquisition. Thus, it was important that there be some
rapport and familiarity between myself and the students akin to the actual teacher-
student relationship.
The selection of this school site also made sense because there was a
mismatch between its minority language students and the language arts program in
use (OCR), which was designed for native English students. The program places
strong emphasis on phonics and assumes that children enter into kindergarten as
native English speakers.
I also had very personal reasons for choosing my current school. This goes to
my original motivation for entering into a masters program. I want to do something
to improve my students’ academic achievement.
44
Program Context
Overview. For the primary grades, the focus of this study, the English
language arts curriculum is primarily the OCR program. During kindergarten through
third grade, the focus is on phonemic awareness and systematic instruction of sound-
spelling patterns of letters and sounds (Bereiter et al., 2000). In addition to daily
phonics practice, students read decodable books (stories with controlled vocabulary
to practice particular sound-spelling patterns), receive mini-lessons in areas such as
grammar, the writing process, text structure, and building background knowledge for
the readings. They also read from a literature anthology (collection of abridged
readings, several of which are group together into thematic units, some of which are
related to science or social studies).
In the fourth grade, OCR shifts emphasis from phonics and phonemic
awareness to word study. The anthology readings shift to a mixture of narrative
(story) and expository (informational) text with increasingly high concentrations of
academic language. The school district mandated 2-1/2 hours per day for OCR,
although anecdotal teacher comments to me indicated up to 3 hours per day to
complete all the OCR mini-lessons and activities. With regards to grammar, a given
mini-lesson typically consisted of a whole group direct instruction lesson, then
guided practice using worksheets, followed by opportunities to extend the lesson into
the students’ own writing work.
One of the primary OCR measures of student progress is reading fluency.
Reading fluency is defined as the number of correctly read words during one minute
45
(total words read minus mistakes). One minute reading fluency scores are an integral
element of OCR periodic assessment throughout all grade levels. This is important
within the context of second language learners. As noted above, Foorman et al.
(1998) judged OCR to be successful based upon a very narrow definition of reading
fluency. While it might be reasonable to equate reading fluency with literacy for
native English speakers, the process of reading is in fact a very complex process
(Celce-Murcia & Olshtain, 2005). Absent comprehension of the language encoded
within the printed text, fluency drills simply reinforce the concept that reading and
writing are meaningless tasks devoid of communicative intent.
Independent work time. OCR includes an instructional block of
approximately 20-40 minutes called independent work time wherein most of the class
works independently while the teacher works with flexible small groups of students
to provide differentiated instruction. This is significant given that much of the
criticism of OCR is that the instruction is designed for native English speakers.
Teachers at my school faced obstacles in adjusting instruction (i.e., providing
differentiation) according to the specific needs of the English language learners in
their classrooms, consistent with what I found in my review of the literature on OCR
Moustafa & Land, 2001; Peck, & Serrano, 2002; Achinstein & Ogawa, 2006,
Achinstein et al., 2004).
The Program Appendix of Open Court Reading, Teacher’s Edition (Bereiter
et al., 2000) defines two goals for independent work time: (1) develop students’
ability to work independently and (2) allow the teacher to work with individuals or
46
small groups of children to address their specific needs. Suggestions for meeting the
needs of English language learners while working with the teacher include: pre-
reading selections, pre-teaching vocabulary, and engaging them in small group
interactive discussions.
This instructional block of independent work time was of particular
significance for this study. It represents an ideal time slot to provide differentiated
instruction and thus provided a context for my research questions. During my
program of study I encountered research regarding the benefits of interaction
between native and non-native speakers. While the original support for TBI had been
based upon the ideal of interaction between native speaker and non-native speaker
(Long, 1996), the actual reality seems to have fallen short of that ideal (Pica, 2005;
Swan, 2005). Indeed, the tension between the constraints of day-to-day practice and
the theoretical ideal seems to be a common current in the literature. Critics of
Krashen’s (1982) input hypothesis found that input alone is not sufficient for second
language acquisition (Long, 1996; Swain, 1985). This raises the question of what
exactly constitutes sufficient comprehensible input, which has not been addressed by
the research. Nevertheless, while an ideal level of comprehensible input is realistic
for a child’s first language acquisition, a similar amount is may not be for second
language students, given the constraints of the classroom.
Thus, in forming the parameters of this study I took into consideration not
only the theoretical ideal but also what would be feasible in the real world of second
language instruction. The ideal would be an instructional context of 6 hours per day
47
of one-to-one interaction with a native English speaker. This would be possible under
conditions of a dual language immersion program that has a balanced mix of native
English speaking students and English language learners (Krashen & Biber, 1988;
Thomas, & Collier, 2002). Another opportunity would be to reorient the teacher’s
participation in whole group discussions to expand students thinking and develop
academic language (Resnick & Nelson-LeGall, 1997; Cazden, 2001; Haneda, 2004;
Gibbons, 1998).
The third avenue, which formed the frame for my study, was to exploit
opportunities for one-to-one interaction between the teacher (native or native-like
English speaker) and the language learner student (non-native speaker) within the
actual parameters English language teachers work under. The question I posed was
how to replicate the results of Long (1996) given a classroom with 20 or more
second language learners and exactly one native English speaker. The independent
work time block seemed like a perfect vehicle for testing a new type of TBI that
would be consistent with the original research of Long (1996) and address the
concerns of Swan (2005).
Writing seminar. Writing in OCR is taught as a process: prewriting, drafting,
revising, proofreading, and publishing. Bereiter et al. (2000) define the writing
seminar as an opportunity (within the revising section of the process) for students to
discuss their work in progress as well as to share ideas for improving it. The writing
seminar is the single instance within the OCR program that specifically calls for a
one-on-one conference between the teacher (native-speaker, or equivalent) and
48
student (non-native speaker, in the case of English language learners). The goal of
the writing seminar is not to teach the student what to write but how to revise.
Besides students meeting in small groups to share ideas about their work, Bereiter et
al. recommend the teacher holds individual conferences with the students. It is a time
for the teacher and student to review student comments, ask questions to clarify
student understanding of the revision process, provide encouragement, and assist the
student in formulating an action plan for revision following the conference.
This is consistent with a large body of research in teaching of the writing
process in general. Specific to writing for second language acquisition, Ferris’ (2002)
own experience and research indicates that teacher involvement in the writing
process should be a temporary support that needs to be phased out as the English
language learner is taught to independently self-edit and self-correct writing pieces.
Teaching the ability to self-correct independently is certainly a laudable goal and
important for first and second language learners alike. However, the underlying
theory and research base of constructivist pedagogy indicates that only through
interaction and assisted performance can new concepts and ways of thinking can be
successfully assimilated into the student’s existing knowledge base (Vygotsky, 1978;
Vygotsky, 1986, De Lisi, & Golbeck, 1999; Beveridge, 1997; Gallego Codes, 2004).
Furthermore, second language pedagogy relating to comprehensible input (Krashen,
& Terrell, 2000), interaction (Long, 1996), output (Swain, 1985), TBI (Ellis, 2003),
and cognitive language learning (Skehan, 1998) all point to the need for the English
language learner to encounter and interact with valid exemplars of standard English.
49
Beck, McKeown, & Kucan (2002) have done extensive research into the
benefits of a print-rich learning environment. The counterpart to this within the
writing process of second language learners would be to make the writing conference
an oral-rich encounter specially tailored to each English language learner’s ZPD as a
English language learner.
Sentence Lifting. After revising, the next step in the writing process is
proofreading (Bereiter et al. 2000). One critical aspect of the writing process is for
students to notice the errors they make repeatedly so as to not make them in the
future. Noticing of errors is solidly based in Piagetian constructivist pedagogy (De
Lisi, & Golbeck, 1999) as well as TBI/FFI literature (Mackey, 2006; Izumi, Bigelow,
Fujiwara, & Fearnow, 1999; Skehan, 1998; Schmidt, 2001). Also, Ferris (2002)
looked at structured methods to provide feedback to English language learners so as
to notice their errors and thus decrease the frequency of their occurrence.
Within OCR, students are given a proofreading checklist and taught to
understand proofreading marks. Then the teacher explicitly models proofreading
through sentence lifting. The teacher takes several sentences from student work and
models how to make the corrections to the whole class. Students are expected to take
responsibility for their own proofreading, to the extent that they have learned various
points of grammar and syntax.
The program context of these English language learner students places
emphasis on becoming independent writers. Since it was designed for native English
speakers, there is an implicit assumption that the students already have a strong
50
command of the English language. Research and theory within second language
acquisition states that it is through interaction with native speakers that second
language learners improve their proficiency (Ammar & Spada, 2006; Long, 1996;
Long, Inagaki, & Ortega, 1998; Norris & Ortega, 2000).
Participants
Researcher participant
Because I had been working as a teacher at this school site for about five
years during the data collection phase of the study, I had an insider’s (emic)
perspective of the program and the specific needs of the students at this school site.
The concept of emic (insider’s) versus etic (outsider’s) perspective and knowledge
comes from cultural anthropology (Morey & Luthans, 1984; Harris, 1979).
I was sole provider of the study’s experimental methods. I held a clear
teaching credential from the state of California for K-12 students with mild to
moderate learning disabilities. I had a certificate in cross-cultural language and
academic development, which is designed to support teachers who instruct English
language learners not receiving instruction in their primary language. I was a
candidate for a masters in teaching English to speakers of other languages (TESOL)
from a major urban university. I had seven years experience working with both
general education and special education students, ranging from 4 to 14 years of age. I
had worked at two schools: the school site (kindergarten through eighth grade) of the
present study from which the participants were selected, as well as another school
site with similar demographics. During six of my seven years of teaching, I used
51
OCR as the major component of the English language arts curriculum. Since the
study’s activities were tied to their OCR lessons, I was able to better simulate the
teacher’s ability to integrate the independent work time activities with the regular
lessons.
Teacher participants
The other teacher participants were the general education teachers of the
students selected for this study. As the experimental methods were connected to the
English instruction the subjects received during the regular school day, they were
involved in the study tangentially. Aside from participating in a post-intervention
interview, they were not involved in the study.
Learner participants
The ten participants of this study were selected from a single public
elementary school, a feeder school for a large inner city urban high school of
California that experiences a high rate of high school dropout. They are referred to in
this study as Participant A through Participant I (nine students, the data for the 10th
student was excluded from the data because she could not converse in English).
Based upon the participants’ English language development level, the study recruited
primarily students of intermediate language development who had been receiving
English language instruction for a period of four years. The first language of all the
participants was Spanish. Their ages ranged from 8 to 9 years old. This age is
significant because it represents students at an intermediate level of English language
development based upon years of English language instruction. In the fourth grade
52
there is a transition in the curriculum from learning to read to reading to learn along
with a change in emphasis to cognitive academic language. Thus, it seemed
appropriate to begin the intervention with students who would be shortly be
transitioning to more cognitively demanding language forms.
Potential learner participants were selected from a pool of two third grade
classrooms (about forty students). There were five participants from one class and
five other participants from the other class.
Recruitment
Ethical Considerations
As participation was voluntary, I felt obliged to provide the participants with
a token of my appreciation for sacrificing their after-school free time to enable the
study to proceed. At the same time, it was important that the value of the gift not
have a coercive effect on them. As the study focused on written language,
participants were given a Lamy Safari fountain pen with a converter and a bottle of
ink (valued at about $35). Participants were able to use the pens during the study and
keep them when the study concluded. I also gave them an after-school snack, a
granola bar for each session which they participated.
I also made myself available for homework assistance at the conclusion of
each 45 minute session. During the initial sessions, several participants had concerns
that they would not have the time to complete their homework because of the time
involved with doing the after-school intervention (especially because several of them
were concurrently enrolled in an after-school program that provided homework
53
assistance). In response to this concern, I remained for an additional 45 minutes after
each session to assist them with homework. About four participants availed
themselves for this additional assistance, although not consistently. Those who
requested additional assistance varied between one to five participants each day.
Participant Recruitment
Approximately one month prior to beginning the study, I spoke briefly to the
students of the two classes and explained in student language that I had to work on
my own big homework project involving writing and learning English and that I was
going to need a handful of students to help me with it. I also told them that helping
me with this project would not affect their grades nor influence decisions about
promotion to the next grade. I showed them the fountain pen and told them that they
would receive one for helping me with my project even if they didn’t continue all
they way to the end. They were given a parent flyer, reiterating the overall topic of
the study, the time investment that would be required (approximately 20 hours),
participant rights to privacy, voluntary nature of the participation, and that no
guarantee of benefits to the participants could be made.
After their parents expressed interest in participation (by signing the parent
flyer), a meeting was scheduled with the student and parents. The students, their
parents, and the two teacher participants were informed about their rights to privacy,
the voluntary nature of participation in the study, and that no guarantees could be
made as to improvement in writing nor English language development.
54
My original intent was to only have 4-8 students participate in the study.
However, it ended up that ten students expressed interest in participation, and so I
decided to include all ten in the study rather than exclude anyone.
Consent forms were developed in collaboration with the Institutional Review
Board approval process preliminary to beginning the study. The primary
considerations in designing the consent and information forms were that the
participants would be informed of their rights and potential risks involved with
participation in the study. The centerpiece of the information meeting was the
consent form. They were given an opportunity to read through the document, given
to them in advance of the meeting, as well as engage in an open-ended dialogue
regarding any other concerns they might have.
Task-Based Writing Instruction
Introduction
TBwI did not exist in the literature, per se. Rather, it is an adaptation of TBI/
FFI and served as the instructional treatment for this study. Unlike many other
examples identified within the literature review, TBwI is essentially a purposeful oral
language exchange and jigsaw task centered on clearly understanding a piece of
student-created written language (hence task-based writing instruction).
Vygotsky (1986) likens written language to a conversation with a blank piece
of paper, lacking both the expressive qualities of oral speech and a live interlocutor.
He sees it as a highly abstract form of communication that does not repeat the
development of speech. TBwI not only uses written language as a platform for oral
55
language exchanges, but it also acts as a scaffold to reduce the cognitive complexity
of the writing task by providing a live interlocutor.
Three Step Process
TBwI of the present study is a three step process. First, the student is given a
writing prompt as a response to literature. This topic sentence identifies some broad
theme within a story likely to be suitable for making a text-self connection. The
student is given a choice of three possible topic sentences of a paragraph and asked
to choose one and write four additional supporting sentences. An example is listed
below.
Quote: “The robbers jumped up when they heard that frightful noise, thinking a
ghost was coming in, and they ran out into the forest in terror.”
(Roxaboxen, p. 267)
Prompt: Choose one of the three sentences. Write it on your paper and add four
detail sentences.
Choice1: I remember a time I was super scared.
Choice2: Sometimes it is hard to be brave if you have too much imagination.
Choice3: When I grow up, I will teach my kids to control their imagination.
Second, the student and teacher sit down together and talk in order to make
meaning of the writing (5-10 minutes). During this time, the other students in the
class work independently or in cooperative groups on writing activities. The English
language learner is the author and the native English speaker (teacher) is the reader.
As each sentence is read, a mini-discussion ensues to identify possible disconnects
between the writer’s intended meaning and the reader’s actual understanding. The
reader might initiate a series of conversational turns with I don’t understand what
56
you are trying to say. In this sense, it is a true jigsaw task because the student has
experiential expertise while the teacher has linguistic expertise. Once there is
harmony between the author’s intended meaning and the reader’s understanding, the
two work together to recast the sentence from the English language learner’s
interlanguage into standard English to effectively convey the intended meaning.
Finally it is written down, and the two continue with the next sentence. An example
is listed below:
NS = native speaker, what I said.
NNS = non-native speaker, what the participant said.
NS: read me the next sentence.
NNS: but I like my old friends [the topic sentence was I remember a time I tried
something new and different, writing about her first day in a new school
away from her friends].
NS: but I like my old friends too?
NNS: yes
NS: I don’t understand. are you saying that you want to keep your old friends?
or you want to make new friends but you also want to keep your old
friends? I don’t understand.
NNS: It’s because, I don’t like to be somebody’s friends. I only like my old
friends.
NS: So you didn’t want to have new friends, you just wanted to keep your old
friends?
NNS: Yeah
NS: Okay, so let’s change the sentence to say But I wanted to keep my old
friends instead of making new ones.
Third, at the conclusion of the writing conference, the student leaves with the
original rough draft (written interlanguage) and final draft product (recast written
standard English) of the TBwI. The student then compares the two written versions
and makes editing marks to the original rough draft by comparing the two. Then the
final draft is recopied by hand. This gives the student a second opportunity to notice
57
differences between his/her written interlanguage and the recast version in standard
English. An example of these two versions is listed below:
Original: I remember a time I tried something new and different. I was eight years
old. Me and my dad went to the building. My dad said there’s different
stuff in the building. You will have fun with new friends. But I like my old
friends.
Recast: I remember a time I tried something new and different. I was eight years
old. My dad and I went to the new school. My dad said, “There’s different
stuff in the building. You will have fun with new friends.” But I wanted to
keep my old friends instead of making new ones.
TBwI as a Focused Task
Ellis (2003) indicates that TBI can be structured to encourage the use of
certain linguistic features, and thus become the subject of instruction (focused tasks).
He contrasted this with unfocused tasks, which are not designed to induce the
language learner to use a particular linguistic feature. Long & Robinson (1998)
characterized focused tasks as Focus on FormS and unfocused tasks as Focus on
Form, in which the task was structured without a specific linguistic focus. In the
present study, the greatest barrier to meaningful communication often was not a
targeted linguistic feature but some other feature that created the interference to
communication. Thus, the principle of primacy of meaning often overrode
considerations of targeted linguistic feature. At times the TBwI was consistent with a
focused task because the targeted linguistic feature (e.g., use of the simple past tense)
did in fact create the greatest interference to communication. Other times it was more
consistent with the characteristics of an unfocused task because it was reactive.
58
Targeted Linguistic Features
Introduction
The tasks were structured such that three linguistic features were targeted for
instruction: (1) using correct verb tense; (2) writing complete sentences; and (3)
writing coherent paragraphs. There were two broad interrelated rationales for
targeting these features. First, the California English language arts standards for the
third grade target these specific features. Second, these three features represent
different ends of the grammar instruction continuum.
On one end of the continuum, traditional English grammar pedagogy
emphasized intrasentence issues such as the ability to correctly conjugate a verb,
given the sentence, verb infinitive, and a given tense (Krashen & Terrell, 2000).
More recent grammar pedagogy has emphasized inter-sentence issues; Celce-Murcia
& Olshtain (2005) stressed the importance of inter-sentence grammar, such as the
ability to write sentences that flow and are consistent within the context of the
paragraph wherein they are placed. My review of the literature did not uncover
previous examples of TBI/FFI investigations specific to these issues.
Standards-based Linguistic Features
Instruction within the California public school system is guided by state
standards, goals for student achievement at the end of each school year. The
Reading/Language Arts Framework for California Public Schools: Kindergarten
Through Grade Twelve (Reading/Language Arts Framework, 1999) lists 54
standards for the third grade, broken down into the strands of reading, written
59
language, listening, and speaking. One of those standards is to create a single
paragraph that develops a topic sentence and supporting sentences that provide facts
and details. A second standard is to identify and use subjects and verbs correctly
while speaking and writing simple sentences.
Given that the participants were working at or below grade level, in part due
to their level of English language development, these were the features they were
exposed to during the direct instruction and guided practice of their regular
classroom. Based upon their grade and level of English language development
(fourth year of English immersion instruction), these features were posited to be
within the participants’ ZPD.
Intrasentence Linguistic Features
Ellis (2006a) examined 17 grammatical structures in terms of explicit and
implicit knowledge. Certain structures had a high performance differential between
the two. Thus for certain structures a student could acquire explicit knowledge
(through grammar mini-lessons, for example), and yet have difficulty in transferring
the knowledge to situations that require implicit language usage (usage in real time
when speaking or writing). Use of regular past tense (-ed) ranked fourth hardest
(only three other structures had a higher score) in terms of transferability of explicit
to implicit knowledge. A high ranking meant that a student would score high on a
test of explicit knowledge (language learning) and low on a test of implicit
knowledge (language acquisition).
60
Given that verb tense tense is a difficult concept to transfer from language
learning (grammar mini-lessons) to real-time usage, it seemed highly suited to an
experimental treatment that targets implicit language acquisition. Therefore, one
aspect of the TBwI was to provide the participants with focused tasks which would
require this grammatical structure.
Intersentence Linguistic Features
The third targeted feature was be the ability to write paragraphs with
coherency wherein the various sentences within the paragraph work in concert to
develop the main idea. Celce-Murcia & Olshtain (2000) stressed the importance of
written text as serving a communicative function. It is possible to write an essay
composed of grammatically correct sentences that do not serve a communicative
function because the reader does not understand the underlying message. Coherency
applies the concept of Grice’s (1975) maxims (quantity, quality, relevance, and
manner) to written language. In the context of this study, it is the degree to which a
topic sentence is developed and supported with with sufficient but not excessive
information so that the written text can be an effective tool of communication.
Procedure
Instructional Intervention
The instructional intervention consisted in 20 after-school sessions of 20-45
minutes, spread out over 4 weeks. The work product of each weekly instructional
cycle followed the following steps: (1) I distributed the weekly writing prompts to
the participants and initiated a brief whole-group discussion to ensure that they
61
understood my expectations and the writing prompts; (2) the participants then
independently wrote a first draft paragraph based upon the writing prompt; (3) once
the first draft was completed we held a writing conference of about ten minutes;
making corrections real-time on my laptop computer, the participant walked away
from the conference with both the rough draft and a computer printout of the final
draft we created together; (4) the participant then made editing marks and corrections
to the rough draft by comparing it against the final draft; and (5) the participant
copied the final draft in his/her own hand.
The procedure generally followed these phases. Teacher-student interaction
primarily took place during the individual writing conferences, except that at the
beginning of each week’s intervention, we had a group discussion of that week’s
writing prompt. The other participants were expected to work independently while I
was engaged in the writing conferences.
Data Collection
Introduction. In order to gain a better understanding of the relationship
between TBwI, FFI, and language learning within this context, the study used a
range of qualitative research methods. In constructing the design, I assumed a
complex medley of roles for myself: participant observer, action researcher, and
practitioner researcher.
The first element of the data collection process was the actual writing samples
of the students. These consisted in the first drafts and final drafts that the students
wrote in their own hand.
62
The second element in the data collection process was to create a clear picture
of what actually occurred during the writing conferences. To do this, an audio
recording of each writing conference was transcribed and entered into a spreadsheet.
The third element in the data collection process were interviews. The original
intention had been to conduct pre- and post-interviews of the students, however the
data collected from the students did not provide meaningful information due to their
lack of language development and lack of metacognitive understanding. Interviews
of the two general education teachers were conducted approximately two months
after the conclusion of the intervention to see if there had been any meaningful long-
term impact of the intervention.
Work samples. Each participant created at least one rough draft and final
draft. Each participant was expected to complete one per week. Each week the
participants were responsible for writing four sentences to complete the paragraph
prompt (topic sentence) prior to meeting me in the writing conference. Afterwards,
the participant was responsible for comparing the handwritten draft and computer
printout final draft created during the writing conference, making corrections to the
draft by comparing the two, and handwriting a final draft.
Primary data set. An audio recording of each writing conference was made
and transcribed into a spreadsheet. Each verbal exchange consisted in one turn. Since
each writing conference was a sentence-by-sentence discussion, each turn was linked
to one sentence within a given writing conference. This spreadsheet served as the
primary vehicle for the data analysis phase.
63
Interview. At the beginning and conclusion of the intervention process, the
participants were asked a series of open ended questions, as follows: (1) why do
people write; (2) what happens when people don’t understand what you write; (3)
what is the most important part of writing; (4) what is it that you don’t like about
writing; (5) what do you like best about writing, and (6) what are the most important
mistakes to fix when you are making your writing better? This interview did not
provide meaningful information. Participants were only able to provide formulaic
answers amounting to a desire to get better grades in school and make their parents
happy. In addition, approximately two months after the conclusion of the month long
intervention, the two teachers were interviewed about their perceptions of how the
intervention had impacted their students.
Data Analysis
Introduction. The primary data source during the four week study were
transcriptions of 35 writing conferences between the researcher and participants.
Although there was an average of four writing conferences per participant, some
participated to a greater extent than other due to levels of writing fluency and
differential levels of attendance during the 20 sessions. Participant B participated in
seven conferences while Participant F only participated in two.
The turns were classified according to two ways: prompt category and target.
Prompt category was the type of conversation technique used by me to stimulate
language learning. The target category was the particular linguistic feature that was
interfering with the task objective.
64
Prompt category. The first classification was the nature of the prompt. For
this, I used the categories from Lyster (2004). His primary categories were Recast,
Prompt-Elicitation, Prompt-Repetition, Prompt-Metacognitive, and Prompt-
Elicitation. Given the nature of the dialogue, it was necessary to add a couple of extra
categories. The first was Read, this is a prompt that asked the participant to read a
sentence from the writing draft. This was used to initiate a series of turns that
revolved around a single sentence. The second one added later was Statement
because it was found that at certain points I would make observations that did not
involve getting a response from the participant.
There were certain classification issues that had to be ironed out during the
analysis phase. The transcripts required an analysis of about 929 turns spanning 35
writing conferences. In some cases it was not immediately clear which category a
turn should be classified into. But I tried to use a systematic approach to doing the
classifications. In cases where prompts had been attempted and failed, and it was
clear that the participant was called upon to simply repeat what I had said, then it was
classified as a Recast.
In cases where I did not understand the sentence and was attempting to elicit
additional information to expand the sentence, it was recorded as a Prompt-
Clarification.
In cases where I was attempting to have the participant make a connection to
a grammar concept, or in cases where I was attempting to say something like does
this sound right? then it was recorded as a Prompt-Metacognitive.
65
In cases where I was attempting to elicit information from the participant, this
was classified as a Prompt-Elicitation. A key difference between this and Prompt-
Clarification was that with the former we had already established a common schema
and understanding of the intended message, with the latter I was attempting to have
the participant produce language that accurately represented the intended meaning.
This was frequently done by means of a cloze technique (beginning a sentence and
then pausing toward the end to give the participant an opportunity to finish the
sentence) or two choices (is it THIS or THAT?).
In some cases it was not easy to distinguish the different types, but having
one person as the rater helped to ensure consistency in the ratings. I also made
multiple passes of the data set to achieve greater consistency. Another issue that
clouded the coding of each of the turns was that in many cases there was no clear cut
distinction between categories. So, for example, some of the turns involved
combinations of more than one category. It often happened that a particular series of
turns would end with a mini-lecture by me and then be immediately followed up by a
different type of prompt regarding a different grammatical focus. In such cases, the
turn was split into multiple turns so that each could achieve purity of prompt
category.
Target category.
The final version of target category is listed below:
(a) End Marks (M.End) – these turns deal mainly with developing a concept
of what is a complete sentence (or, a complete thought). This was associated with the
66
need to add periods. In some cases no punctuation marks were included, and in other
cases the periods were placed such that it was obvious that the participant didn’t have
a clear concept of what constituted a complete thought.
(b) Other Marks (M.Other) – this was a very rare category. There were only a
few instances of this category. They dealt primarily with apostrophes (indicating
possession) and commas.
(c) Quote Marks (M.Quote) – This dealt the the conventions of indicating
direct speech. It includes the use of an offset comma, quotation marks, and
capitalization within the quotation.
(d) Semantics (Sem) – This was a fairly rare category. It dealt with those
instances of word meaning that interfered with the ability of the participant to
communicate intended meaning. In some cases issues of Semantics were categorized
as Schema (see below) where it was caused by an excessive reliance on
contextualized speech (that does not rely upon exactness and specificity to
communicate a meaning because of shared context and/or schema between the
speaker and listener). In some cases, problems with Semantics had to do with a lack
of understanding of the prompt.
(e) Prepositions (Prep) – this category dealt both with the use of prepositions
as well as phrasal verbs (idiomatic usage of prepositions associated with a particular
verb).
(f) Pronouns (Pronoun) – this category was at times closely related to
Schema, in cases where contextualized speech had habituated the participant to use
67
pronouns without clearly stating antecedents because of a reliance on contextualized
speech. In other cases, the participant confused nominative with objective pronouns.
(g) Schema (Schema) – This category had to do with sharing just enough—
not too much and not too little—information so that the reader would understand. I
suspect that it was due to the participant’s lack of familiarity with decontextualized
forms of discourse. This category tended to be the area of focus with participants
who had greater levels of English language development.
(h) Simile (Simile) – this category occurred rarely, usually when the
participant had an inappropriate usage of a simile that would create confusion for the
reader.
(i) Syntax (Syntax) – this was a fairly rare category that dealt with improper
word order (e.g., I found a dog big).
(j) Topicality (Topicality) – this category was strongly connected to Schema.
In cases of a schema focus, additional information usually needed to be added in
order to create clarity of expression through a shared schema. With problems of
topicality, usually there was extraneous information included in the paragraph that
did not develop the given topic sentence. In some cases problems of Topicality were
related to issues of Semantics because the participant did not clearly understand the
meaning of the prompt. For example, in one case when given a topic sentence I
remember a time I found something by accident, the participant associated by
accident with something bad happening and, thus, proceeded to write about how he
had become injured instead of writing about how he had once unexpectedly found
68
something valuable. In another example, the paragraph was written to the prompt I
remember a time I used my imagination to play, but in fact the participant mostly
wrote about events that took place after he had finished playing with his imagination.
(k) Verb Tense (V.Tense) – this category had to do with whether the verb
tense is consistent with the given topic sentence. For example, if the prompt was to
remember about something that had happened in the past, then all the remembrance
sentences should be in the past tense instead of present tense. An example of this was
provided in the very first excerpt presented at the beginning of this chapter.
(l) Coherency (Coherent) – this category dealt with the flow of sentences
together, the manner in which they were connected. This was one of the finer points
of discussion that we addressed when other issues were not quite so pressing.
Reliability and Validity
This was an exploratory study that sought to apply a method (recasts and
prompts) developed within the context of oral language development to a new
context (written language). Due to the small sample size, it was not expected that
definitive answers would be found to the research questions. One of the issues
revolved around the reliance upon the transcripts of the writing conferences as the
primary data source without triangulation from other data sources. I attempted to use
other data sources, including pre/post interviews and participant’s editing of the
papers. However, these data sources proved to be unreliable due to implementation
issues that will be discussed in the following section. Validity and reliability, to the
extent possible, was achieved through multiple passes of analysis of the data set so
69
that there would be consistency in the manner of coding and classification, and that
the coding/classification system used reflected the meaning-focused nature of the
interactions that took place.
70
CHAPTER 4, RESULTS
Introduction
As was discussed in the previous chapter, my understanding of the data
methods evolved throughout this project. In like manner, my conception of how to
present, analyze, and evaluate the data changed over time. At the conclusion of the
research project, I had approximately six hours worth of transcribed recordings from
35 writing conferences plus about 60 pages of writing samples.
Once I completed the process of synthesizing and coding the raw data (see
the Procedure section of CHAPTER 3, METHOD), I set about to understand how I
could connect this data to my research questions. To do this, I first had to reflect on
my research questions and original motivations for beginning this study.
The second research question regarding differentiated instruction was
inspired by Long’s (1996) investigation into reactive negative feedback and his
interaction hypothesis. He paired up native and non-native English speakers in a
classroom and provided them with opportunities to interact. His research setting
seemed like an ideal linguistic classroom environment of half native speakers and
half percent language learners. Similarly, I wanted to create a rich linguistic
environment for English language learners in an urban school, working under the
constraint of an insufficient proportion of native English speakers.
My personal anecdotal experience and review of the literature had uncovered
a need for differentiated instruction such as reactive negative feedback. I found that
traditional quantitative measures within statistical analysis were not suitable to
71
quantify the degree of differentiation within instruction. Therefore, I set about
through a trial and error process (see APPENDIX E) to ascertain reliable quantitative
measures of differentiation. Once I had broken down the writing conferences into
discrete turns, categorized the type of exchange (e.g., prompt or recast), and the
targeted linguistic feature, I then gauged the quantity of interaction by the quantity of
text transcribed (measured in number of characters). This data is summarized in the
Quantitative Data section. For additional information, please consult APPENDIX A.
The primary inspiration for this study had been the seminal work Thought
and Language (Vygotsky, 1986). This book led to the first question, which was to
understand the issues associated with using TBwI as a platform for communicative
language teaching principles and constructivist pedagogy. To answer this type of
question, it was necessary for me to do an analysis of the transcripts to look for
specific instances of my assessment of ZPD as well as examples of interaction that
gave the participants opportunity to construct new knowledge. The Qualitative Data
section (see below) presents this analysis.
The third research question was to examine how TBwI might impact second
language acquisition. This was the most difficult question to examine from the data
set given the parameters of the study (high degree of differentiated instruction and
thus multitudinous linguistic features address over a relatively short span of time). To
answer this question I examined the teacher interviews and writing conferences; this
information is presented primarily within the discussion of Qualitative Data section.
72
Primary Data Set
The primary data set was the transcript of the 35 writing conferences along
with a concurrent analysis of: the (1) sentences within the rough draft, (2) final recast
sentences of the final drafts, (3) what the participant said, and (4) what I said to
accomplish this. Below is an portion of the Primary Data Set:
Figure 1: Sample Turns within the Primary Data Set
1
2
3
4
5
G H I J Q R
Sentence Recast Teacher Student P Target
When I get older, I want to
go treasure-hunting.
When I get older, I want to
go treasure hunting.
Okay, why don’t you start by
reading the first sentence.
When I get older, I want to go to
treasure hunting ... I could ...
Rea n/a
then I gay som mony to my
grama.
Then I will give some money
to my grandparents.
okay, what’s the next sentence?
I give some money to my grand
parents.
Rea n/a
Then I give some money to my
grand parents ... Wait a minute ...
wait a minute ... wait a minute ...
You are going to give some money
right now?
will Met V.Tense
WILL, that’s right ... you are so
smart ... then I WILL give some
money to my grand parents ...
okay great.
- Sta V.Tense
From this exchange, we can see that there were three turns to recast this
sentence to the future tense. The given topic for this paragraph was to go treasure
hunting in the future so it was necessary for the participant to use the simple future
tense. Excluding the Prompt-Read (Rea), a total value of 193 (total length of the text
string) was used to recast the participant’s sentence from the past tense to the future
tense. This was primarily done by means of a Prompt-Metacognitive (Met) to get the
participant to state will, followed by a Prompt-Statement (Sta) wherein the entire
participant’s phrase was recast based upon changing gave to will give. The last turn
was categorized as Prompt-Statement as opposed to Recast because the participant
was not expected to provide a response and was therefore one-way communication.
73
Quantitative Data
Linguistic Focus
Figure 2 is a summary of the TBwI that took place over the month long
intervention. The percentage of time spent on each linguistic target was based upon a
turn-by-turn classification based upon the targeted type. It was generally a reactive
form of instruction. As each sentence was discussed within the writing conference, I
made a snap decision about what element of language usage posed the greatest
barrier to understanding.
Figure 2: Overall Linguistic Focus of Prompts/Recasts
Verbs
22%
Schema
22%
Punctuation
19%
Topicality
16%
Pronouns
6%
Coherency
6%
Semantics
3%
Syntax
2%
Simile
2%
Prepositions
2%
Other
9%
With regards to question two, differentiation of instruction, this data clearly
shows that there was a high degree of differentiated instruction. As discussed in
CHAPTER 3, METHOD, there were three targeted linguistic features: (1) using
correct verb tense; (2) writing complete sentences; and (3) writing coherent
74
paragraphs. Correct verb tense was within the category verbs (19% of turns). Writing
complete sentences was a subset of the category punctuation (punctuation: endmarks
was 12% of turns). Writing coherent paragraphs was within the categories of
topicality (16% of turns) and coherency (6% of turns). While these three items were
certainly addressed during the writing conference, roughly half (53 percent) of turns
addressed non-targeted linguistic features.
Comparison of the Overall Linguistic Focus with those of the individual
participants (Table 1) provides additional evidence that there was a high degree of
differentiated instruction. Please refer to APPENDIX A for additional information.
Table 1: Linguistic Focus of Prompts and Recasts
Student Participants
FOCUS Overall A B C D E F G H I
End Marks 12% 5% 1% 12% 31% 26% 8% 8% 21% 6%
Marks, Other 1% - - - 9% - - - - -
Quotation 6% - 5% 18% 9% - 7% 17% - 4%
Semantics 3% - - 19% - 5% 16% - 4% -
Prepositions 2% - 2% 7% 2% - 2% - 2% 1%
Pronouns 6% - 6% 6% 6% 6% 10% - 22% 9%
Schema 22% 16% 50% 22% 27% 5% 10% 15% 11% 13%
Simile 2% 16% 3% - - - - - - -
Syntax 2% 7% 2% 6% 3% - - 3% - -
Topicality 16% 3% 11% - 11% 26% 40% - 21% 30%
Verbs, Other 1% 1% - 6% 3% - - - 3% -
Verb Tense 22% 53% 2% - 8% 21% 2% 56% 18% 36%
Coherency 6% - 18% 4% 7% 9% 5% 1% - -
75
For example, the linguistic category verbs was a targeted linguistic element
and was the overall linguistic focus 22 percent of the time. However, the actual range
was from 0 to 56 percent.
Prompts vs. Recasts
The figure below shows the relative use of prompts versus recasts for the
Participants (A-I). The percentage was calculated by dividing the number of Prompt
turns by the sum of Prompt plus Recast turns. For example, Participant A had a total
of 36 Prompt turns and 6 Recast turns (36/[36+6]=.86).
Figure 3: Ratio of Prompts to Recasts
86%
100%
94%
98%
94%
90%
81%
84%
82%
75%
80%
85%
90%
95%
100%
105%
A B C D E F G H I
Student Subject
Percentage of Prompt Turns (versus Recast)
As can be seen, there was a marked bias for the use of prompts. In essence,
recasts were used as an conversation tool of last resort. Prompts were used as tools to
coax the participant into recasting the sentence into standard English. When this was
76
not possible or feasible, I recast the sentence myself into standard English and gave
the participant the opportunity to repeat the recast sentence. Using this criteria,
prompts were more explicit while recasts were more implicit instruction.
This data addresses Question Two, degree of differentiation of instruction.
Clearly, some participants were able to handle more cognitively challenging
interaction (prompts). Other participants were challenged to operate within the outer
edge of their ZPD, which required me to switch from prompts to recasts to provide
additional scaffolding.
Turns vs. Sentences
Figure 5 shows the average number of turns to recast each sentence. It was
calculated by taking the total number of net turns (excluding turns prompting the
Figure 4: Ratio of Turns to Sentences
2.8
2.4
2.5
3.6
3
2.7
4 4
5.1
0
1
2
3
4
5
6
A B C D E F G H I
Student Subject
Ratio of Turns/Sentences
77
reader to read from the draft) divided by the number of sentences (excluding the
teacher-created topic sentence) that were successfully recast into standard English.
For example, Participant B had 94 turns to 40 sentences (94/40=2.4).
This data along with the previous set gives a sense of the degree of difficulty
in guiding the student through this process. Participant B required the fewest number
of turns to recast his sentences into standard English because of his level of English
language development. Participant H had a very low level of language development
and required additional turns. Participant I had the highest number of turns because
she had the greatest difficulty in following the pragmatic element of conversation.
This data addresses Question Two, degree of differentiation of instruction.
Clearly, some participants required additional scaffolding in the form of additional
conversational turns of teacher talk to successfully recast a given sentence from the
student’s interlanguage to standard English. This additional scaffolding was provided
in varying degrees (a ratio from 2.4 to 5.7, indicating average number of turns to
recast a given sentence from the student’s interlanguage to standard English)
according to the participant’s needs and the flow of any given interaction. It provides
additional indication of differentiated instruction.
Teacher vs. Student Talk
The below data shows the amount of teacher to student talk during the recast
and prompt turns.
As stated previously, the amount of talk was quantified by taking the text
length within each turn that was transcribed. This measure helps to give some
78
Figure 5: Ratio of Teacher to Student Talk
6.2
5.1
8.5
10.1
7.8
15.2
8.3
6
3.5
0
2
4
6
8
10
12
14
16
A B C D E F G H I
Student Subject
Ratio of Teacher/Student Talk (Prompt & RC)
understanding as to whether the writing conferences were successful as an efficient
tool for language instruction. For example, Participant B had a high level of
language development and required fewer turns per sentence to recast the sentences.
In contrast, Participant G had a very low level of English language development, and
thus, I needed to talk quite a bit more (e.g., using many closed-ended questions) to
make meaning of her writing.
This data addresses Question Two, degree of differentiation of instruction.
Clearly, some participants required additional scaffolding in the form of higher
quantity of teacher talk to successfully recast a given sentence from the student’s
interlanguage to standard English. This additional scaffolding provided in varying
degrees (ranging from a ratio of 3.5 to 15.2, greater number indicating relatively
79
more teacher talk) according to the participant’s needs and the flow of any given
interaction indicates that there was a high degree of differentiation of instruction.
Qualitative Data
Introduction
The qualitative data consisted primarily in an analysis of specific passages of
the transcribed writing conferences, the teacher-created writing prompt (topic
sentence), rough draft, and final draft. Interviews were also attempted with both the
participants and teachers; most of that data was of limited usefulness partly due to
my inexperience in conducting interviews and gathering data from these sources.
From the interviews and questionnaires of the participants, no conclusive data was
acquired. The teacher interviews did yield useful information about academic growth
perceived by the teachers.
Interviews
Question Three explored the impact of TBwI on language acquisition.
Interviews with the general education teacher of the participants were conducted at
about the end of the school year (the interventions had taken place roughly mid-
year). Both teachers indicated that the participants’ interest in writing and the writing
process had either remained the same or had increased as a result of the intervention.
The first teacher found it difficult to comment on the impact of the study upon
the participants: “Well, to tell you the truth, I had a very strong writing program
myself this year. And I’m having trouble really distinguishing your impact versus
mine. They did really improve, but everybody in the class did too.” In contrast, the
80
second teacher observed a marked improvement with the participants’ writing. But
not all participants improved their writing to the same degree or in the same manner.
She indicated that some participants improved their writing fluency, as the following
excerpt suggests:
[Participant D] would always kind of linger with his thoughts about what he
would write. That’s not a problem to think before you write, but he had long,
sustained periods of thinking before he would write, or he would not want to
write. So, specifically for him, he narrowed down that thinking time and
actually produced. He would be able to pair/share with somebody else and
talk to them about what his topic was going to be and then get right to it.
With other participants she noticed an improved ability to edit and revise,
possibly due to a greater awareness of the communicative intent of writing:
With [Participant B], I noticed that he was able to go back and re-read his
work and check for grammar and notice that if something was written in
incorrect academic English. He would try to rephrase it. And so I often
noticed that he would go back to proofread.
With some participants, she observed changes in actual sentence writing,
possibly as a result of noticing differences between her interlanguage and standard
English or perhaps simply due to a greater awareness of audience:
[Participant C] used to write with lots of run-on sentences, and sometimes
her sentences would make no sense, or she wasn’t producing what she was
trying to communicate. And after a process of finishing her work and going
back and listening to others read her work, then she would stop herself and
say, ‘Okay, I know what I need to do.’ when she heard other people read her
work.
Some of the participants seemed to increase their ability to write more
cohesive paragraphs, as the first teacher pointed out in this section of the interview:
[Participant I] has always been very verbal, but she would get easily off the
track and [do] anything to keep talking, on paper or in person. Her writing
81
became more focused and structured. When she seemed to learn the formula
or the knack for putting a main idea and listing some details. She got better at
that, instead of just rambling on and on and on.
In contrast, the first teacher noted that Participant A had made a lot of
improvement in his ability to write more effective sentences.
[Participant A] saw a lot of improvement this semester ... he seemed to be
able to write more. His sentence structure seems to be a little more complex,
using commas in a series, making longer sentences. He struggles so much
though, orally.
While teacher interviews were not able to quantify the impact of TBwI on
second language acquisition (Question Three), they do indicate that there was a
perceived impact on second language acquisition some months after the conclusion
of the study.
Primary Data Set
Introduction. As mentioned above, the key elements of TBI are: (1) authentic
language use; (2) primacy of meaning; and (3) a communication problem to be
solved. An analysis of the transcripts and writing produced many instances that
satisfied these criteria. While a traditional writing conference would be focused on
training the English language learner to independently write and proofread, the TBwI
is focused on conversational interaction that recasts written interlanguage into
standard English. The analysis also indicated consistency with constructivist
pedagogy (Question One).
TBwI as a jigsaw task. In some cases the problem involved something simple
such as correct verb tense usage, as the excerpt below illustrates. Participant A was
82
writing about a time he found something special. There were two problems with this
sentence. First, he used the simple present tense then I find a dog sad to refer to
something happening in the past, which could create confusion for a reader. Second,
he used Spanish language syntax for adjectives (a dog sad for a sad dog). Instead of
making the correction for him so that all he would have to do is recopy the sentence,
I confronted him with the reader’s confusion; then I invited him to supply the correct
answer himself, which he did. In the last exchange I use stress/intonation to offer him
the two choices and invite him to self-correct the mistake.
Draft = the original sentence from his draft
Recast = the original sentence recast into standard English
NS = native speaker, what I said.
NNS = non-native speaker, what the participant said.
Draft: then I find a dog sdu
Recast: Then I found a sad dog.
NS: Okay, then I ...
NNS: find a dog.
NS: Then I find a dog ... wait a minute! you FIND a dog right now?
NNS: found
NS: found, very good ... Then I FOUND a dog.
NS: Okay, so then I found a dog SAD ... or then I found a sad DOG.
NNS: a sad dog.
--Conference 08 / Turns 22-27,
This exchange satisfies all the criteria of TBI and is a genuine jigsaw task.
First, he was making a personal connection to a piece of literature. They had been
reading about a story character who went treasure hunting and was invited to talk
about a time he had done something akin to that. I did not ask him to create a
paragraph for the purpose of making red marks on it; instead, he was communicating
something of interest to others. Second, there was a primacy of meaning. I showed
83
genuine interest in learning about the episode he read to me. When I made grammar
corrections, it was to help the reader (me) to understand what he was trying to say.
There was an occasional focus to grammar, but only within the context of trying to
understand his story. Third, there was a communication problem to be solved. As the
native English speaker, I had expertise in using written English to effectively convey
ideas from writer to reader. Because he was recounting a personal experience, he had
expertise about what actually happened that day he found something special.
Working together we did a true jigsaw task.
TBwI as a focused and unfocused task. Notice that the above passage also
highlights the flexibility of this form of instruction: the ability of the teacher to plan
for certain important topics. The participant was given a choice of the three
following prompts and told to write four additional sentences to make a complete
paragraph:
Choice1: I remember a time I found something special.
Choice2: Sometimes you can find something special by accident.
Choice3: When I get older, I want to go treasure-hunting.
By constraining him to use a single type of verb construction (simple past,
modal can, modal want), I had created a planned opportunity to provide implicit
instruction in verb tense usage. I also planned for additional aspects to focus on.
Specifically, I wanted to make sure that the participants would have a clear
understanding of a complete sentence. As the participant sat down to talk with me, I
readied my laptop computer to retype and correct her rough draft based upon our
conversation together. My rule was that we would only talk about one sentence at a
84
time. By reading the story in chunks of one sentence, this aspect of the task became
focused towards teaching the concept of a complete sentence:
NS: okay, so what is the next sentence?
NNS: I was eight years old me and my dad--
NS: okay, I was eight years old ... that’s really like one sentence, right? ... okay,
so let’s call that a sentence ... okay, what does the next sentence say?
NNS: me and my dad went to the building it was--
NS: okay, so that’s the next sentence, right?
NNS: yes
NS: so, me and my dad went to the building ... just one thing, I notice here ...
ME go to school, does that sound right? ... ME go to school
NNS: no
NS: it shouldn’t be ME went to school, it should be ...
NNS: I went to school
--Conference 06 / Turns 3-8
Notice that in this conversational exchange, as with the previous example, it
displays both focused (planned) and unfocused (reactive) elements. By stopping and
interrupting the participant as she was reading, I was providing implicit instruction
about what is a complete sentence. Also, I had a plausible rationale for having her
stop: I was typing her paragraph into my computer one sentence at a time. Unlike
Participant A of the first example, Participant I had no problem with verb tense
usage and adjective placement, but she had an unclear understanding of pronouns
and nominative/objective case (she wrote me and my dad instead of my dad and I).
This was something I had not planned on teaching, but since it came up in her
writing and posed the greatest potential confusion to the reader, I chose to address
this linguistic feature. Whether we addressed linguistic features I had planned for or
simply problems I reacted to as they arose, those choices were always made in the
85
context of making the paper easy to read; thus primacy of meaning was consistently
maintained during the TBwI.
TBwI and constructivist pedagogy. One key aspect of constructivist
pedagogy is that the student has to take ownership in the learning process, that she
must actively construct new knowledge and then incorporate that within the schema
of previously learned information.
Below, Participant A is writing about a future event: going treasure hunting
when he gets older. He attempted to use the modal could as an alternative future
tense construction. However, I wanted him to use maybe I will to improve his ability
to write with the simple future tense. He was confronted with a cognitive failure (the
use of the modal could is not an acceptable substitute for the future tense) and
attempted to construct new knowledge. Confronted with cognitive failure, he
hypothesized that by combining already with went could be a substitute future tense.
Draft: I cout find som mony.
Recast: Maybe I will find some money.
NS: huh, there is one thing I don’t understand about this. Are you going to go
treasure hunting right now?
NNS: no, when I get older
NS: So, I usually say I could, I could find some money or I could do this or I
could do that when I’m talking about something right now. Are we talking
about something right now?
NNS: no
NS: okay, so how can we say that, not talking about today but about the future?
how can we say that? ... instead of ‘I could’, we could say ...
NNS: already went to treasure.
NS: okay, but instead of saying I could, we could say maybe I will, okay?
maybe I WILL find what would we say, maybe I ...
NNS: I will go.
--Conference 01 / Turns 5-6
86
TBwI and targeting the ZPD. The ZPD represents what is possible for the
student to achieve under conditions of assisted performance. Notice that in the above
example, real-time assessment information is used to guide instruction and determine
the student’s ZPD. In order to recast this sentence into the simple future tense, I
began with two turns of metacognitive prompts, attempting to stimulate recall of the
simple future tense. Then I did an elicitation prompt, at which point he formed an
incorrect hypothesis. Then I realized that I had not given him sufficient scaffolding to
achieve this objective. Therefore, I switched gears and recast the sentence by saying
maybe I WILL find, to which he responded with I will go. He did not read the
complete sentence but I had accomplished my objective: under conditions of assisted
performance he was able to use the simple future tense.
The next excerpt represents a prime example of using TBwI to create
rigorous instruction that pushes the envelope of second language acquisition. It also
demonstrates second language acquisition occurring in real-time. In this example,
Participant F was writing about how she will teach her children to use their
imagination when they grow up. This was a very cognitively demanding writing
prompt because she was not only writing about an abstract concept (imagination) but
also about an event that had not yet happened.
Primacy of meaning was maintained throughout. I had planned the task to
focus on the use of the future tense. She did not write in the future tense as I had
expected. However, it immediately became clear that she did not understand the
concept of imagination. If she did not understand the writing prompt then it would
87
not serve as authentic communication, so I zeroed in on her understanding of
imagination rather than the other issues. Pay especially close attention to the last
conversational turn. Over the course of six turns I provided her with assistance so
that she could construct an understanding of imagination; on the very last turn I ask
her to read the next sentence she wrote in her draft: I drawed a frog. I had just recast
the first sentence to read: my teacher said to draw a picture of a frog. She then self-
corrects and instead of reading the sentence she wrote, she states, I imagined a frog
dancing. This exchange is consistent with Vygotsky’s (1986) conception of writing
as a highly cognitively demanding task akin to conversing with a blank sheet of
paper. Under conditions of assisted performance she was able to actively construct
new knowledge and refine her understanding of cognitive academic language (i.e.,
the word imagination). In the process, we witness in real time her assimilation of
new information in her second language.
Draft: One day my teacher said to draw a imagination. I drawed a frog.
Recast: One day my teacher said to draw a picture of a frog and use my
imagination. I imagined a frog dancing.
NS: okay, what is the next one [sentence]?
NNS: One day my teacher said to draw an imagination.
NS: to draw an imagination?
NNS: to draw a picture
NS: to draw a picture, to draw a picture of an animal?
NNS: an animal, a frog
NS: to draw a picture of a frog ... and what were you supposed to USE?
NNS: huh?
NS: what were you supposed to USE? you ...
NNS: imagination
NS: so one day, my teacher said to draw a picture of a frog and ...
NNS: use my imagination.
NS: use my imagination. okay, and is this telling us what we are going to do
when we grow up?
88
NNS: no
NS: okay, next?
NNS: I imagined a frog dancing. [self-corrects while reading sentence]
--Conference 17 / Turns 23-30
In another example, Participant A wrote about a rock he had found one day.
He described it as like brand new. This did not make sense to me because rocks are
millions of years old; they cannot be brand new. What he was trying to say was that
it looked like something brand new. Therefore, I devoted seven turns to helping him
recast the sentence into It was like a brand new toy. It was something he was
incapable of doing under conditions of independent practice (typical testing
conditions), but under conditions of assisted performance he was able to achieve this
extremely cognitively challenging task.
TBwI as a form of communicative language teaching. What becomes clear
through reading all the transcripts and fully digesting what transpired during that
month is the power of communicative intent. To a second language learner this can
be both a source of frustration and inspiration. When a strong motivation (harnessed
by communicative intent) was present, it was possible to create cognitively
demanding interactions between myself and the participant. The effectiveness of this
constructivist pedagogy seemed highly correlated to motivation.
On the high end of the English language development spectrum was
Participant B. In the following passage he is writing about a time he was really
lucky. He asked his father for three dollars but received four because there was a new
bill that stuck to the other three. What is interesting is that he self-corrects.
89
Draft: He gave me three but when I check to see if he gave me really.
Recast: He gave me three dollars. When I checked to see if he really gave me three
dollars, I saw one more dollar.
NS: okay [read the next sentence]
NNS: he gave me three dollars, but when I checked to see if he really gave me
three dollars.
NS: ah, so you say, When I checked to see if he gave me REALLY ... so you
changed that to be if he REALLY gave me [nss self-corrected while reading
his sentence].
--Conference 16 / Turns 2-3
Participant B displayed no difficulty with the targeted linguistic features.
Nevertheless I remained focused on the communicative aspect of writing. In the
excerpt below he came to grips with the unique aspect of written communication:
lack of a shared context between writer and reader. His sister asked him to do the
milk for his baby sister, but I didn’t understand what he meant. As the reader, I was
struggling to understand his intended meaning. By working together we were able to
make his writing a more powerful communicative tool. Note that TBwI as an jigsaw
task was subtly altered for him; it was not my expertise as a native speaker, but rather
my perspective as the reader and live interlocutor that most benefited him.
Prompt: I remember a time I tried something that was hard to do.
Draft: it was when my sister told me to do milk for my baby sister.
Recast: it was when my sister told me to prepare the milk for my baby sister.
NS: for my baby sister? ... okay do milk is kind of ... so let’s see if we can make
this ... because I don’t quite understand when you say, do milk ... you mean
that you ... what did she want you to do?
NNS: make the bottle of milk
NS: she wanted you to PREPARE the bottle, or she wanted you to FEED the
baby?
NNS: to prepare
NS: ah, She told me to prepare the milk for my baby sister. ... okay, that’s good.
--Conference 24 / Turns 2-4
90
But participants at the lower end of the English language development
spectrum also benefited from this approach, precisely because of motivation.
Participant G participated in five writing conferences, but only three were successful
(i.e., resulted in an authentic communicative exchange and a final draft). The excerpt
below was her first attempt at a writing conference. She wrote about how one can
find something special by accident. When I asked her to read her writing, she could
not. When I offered to orally transcribe her paragraph to me, she could not.
Draft1: the illas sientheri anshe ranche
Draft2: chanacx the rienteu the riendango
Draft3: the man ret sad the sanche and the sanche canchez melisahez chancez and
NS: okay, so what is your second sentence?
NNS: the ... the ...
--Conference 04 / Turns 3-7
Her second attempt at a writing conference also ended in failure. She
attempted to write about how it is important to help one’s parents in hard times,
about a time she helped her parents to paint the house. Somehow the two of us
(reader and writer) got completely lost, thinking that it was dealing instead with wild
animals in ancient times.
Draft: that there o there tan thaki the roero that the romeronchez to the romero that
the carthe thea the cur was hawer Merour Melissa Sanchez wa to hav the
Kerome the carwas Lhero.
Recast: You should help your mom and dad with painting the house.
NS: okay, good, what is the next sentence?
NNS: there ... there ... a leopard ... that ... live ... roman
NS: in the city of Rome or in Roman times?
NNS: roman time
--Conference 20 / Turns 1-10
91
Finally on the third try she was able to perform a successful writing
conference. Just before the writing conference of the excerpt below, I observed her
practicing to read her draft several times prior to sitting down with me, as if she were
trying to commit it to memory. She also enlisted the help of a couple of other
participants. What I display below is not the actual writing conference, which totaled
66 turns, but rather a comparison of the draft to recast sentences. By comparing it
with the previous two samples, tremendous improvement can be seen. It was the
power of communicative language teaching that motivated her to struggle through a
very difficult task.
prompt: When I grow up, I will teach my kids to listen and pay attention.
1draft: my kids was to lafen when the techer sad no lafen pless kids
1recast: My kids will not laugh when the teacher will say, “No laughing please
kids.”
2draft: oke mes sad the kids
2recast: The kids will say, “Okay Ms.”
3draft: the techer sad can to the for saod the techer pless
3recast: The teacher said, “Can you stop please.”
4draft: kids oke techer
4recast: Then kids will say, “Okay teacher.”
5draft: he red a books to the kids.
5recast: Then he will read books to the kids.
--Conference 34
Not all the writing conferences were equally successful, but these excerpts
are reflective of the powerful aspect of communicative language teaching.
Conclusion
In conclusion, the results answer the three questions of this study to varying
degrees.
92
Regarding Question One, TBwI as a form of TBI/FFI and constructivist
pedagogy, qualitative data analysis indicates that TBwI was a jigsaw task that
provided meaning-focused interaction between myself, a native English speaker, and
the participants, English language learners. There was authentic language use and a
definite communicative intent as we worked together to recast interlanguage
embedded within the writing drafts into standard English. Each of the two persons
had a critical expertise that needed to be used in concert. I the teacher had linguistic
expertise of English while the participant had experiential expertise of the actual
text-self connection that was being made.
Regarding the question of incorporating constructivist principles, the two
primary principles focused on were the my ability to work within a participant’s
given ZPD (at both the high and low range) and provide varying degrees of support
that stopped short of explicit instruction to allow the participant the opportunity to
notice differences between the written interlanguage and standard English as well as
the freedom to make linguistic hypotheses and receive immediate feedback regarding
them, thus, creating additional opportunities to assimilate rather than accommodate
new learning.
Regarding Question Two, the impact of TBwI on differentiated instruction,
the data set demonstrates qualitatively and quantitatively that differentiation of
instruction can indeed be measured and quantified. Across a variety of measures, the
data show that in TBwI was effective in providing differentiated instruction to the
participant students.
93
Regarding Question Three, the impact of TBwI on second language
acquisition, the data set was not conclusive. Because the focus of data collection was
on the writing conferences themselves rather than pre- and post-intervention
measures, because of the wide variety of linguistic items targeted, because of a lack
of triangulation of data sources, because of the complex and long-term nature of
second language acquisition, the data set was not able to demonstrate measurable
gains in second language acquisition. However, the data set provides some indication
that TBwI may have impacted second language acquisition for at least some of the
student participants.
In conclusion, the month long experimental intervention was successful in
beginning the process of understanding how TBwI may be a useful form of
communicative language teaching that incorporates principles of constructivist
pedagogy and second language acquisition as well as impact the level of
differentiated instruction in a classroom of English language learners.
94
CHAPTER 5, DISCUSSION
Introduction
The original impetus for this project had been to understand the impact of
incorporating pedagogy of second language acquisition and constructivism into a
classroom of second language learners. I was exposed to the literature from these
fields, and what followed was an investigation about turning theory into practice.
How could I enhance a structured English curriculum that emphasized explicit direct
instruction? How could I incorporate principles of second language acquisition and
constructivist pedagogy? How would this impact the students’ learning? Could I
increase the ability of the classroom teacher to provide differentiated instruction that
would effectively target the upper band of each participant’s ZPD? Would this result
in second language acquisition? These were my questions.
With regards to Question One, understanding the issues involved with
incorporating principles of second language acquisition and constructivist pedagogy
into instruction, the results demonstrate that TBwI is a valid tool for incorporating
these principles into the classroom of second language acquisition.
However, there are no simple and pat answers to the challenges of education.
In search of simple answers, the pendulum of education policy sometimes swings
from end of the spectrum to the other. TBwI is a highly differentiated form of
holistic instruction with an emphasis on reactive negative feedback. In contrast, OCR
is a program that emphasizes explicit whole group instruction. The challenge for
educators is to arrive at a synthesis of often contradictory teaching methods and
95
focus on the needs of the students rather than dogmatic positions. As such, TBwI
may be a useful element of successful instruction.
Question Two asked how TBwI might impact the degree of differentiated
instruction within a mixed ability classroom. Analysis of the data set shows that
TBwI provided a high degree of differentiated instruction. Each participant received
interaction that was uniquely different from the others along the implicit-explicit
teaching continuum. The actual mix of topics discussed was also varied.
Question Three asked how TBwI might impact second language acquisition.
Based upon analysis of the data, personal reflection, and teacher interviews, it
appears that TBwI did impact second language acquisition to some extent for at least
some of the participants. Given the short amount of time and complexity of the
language acquisition process, it was difficult to quantify this.
Q1, Using TBwI in the Classroom
The first research question was to identify the issues associated with using the
writing conference as a platform for communicative language teaching consistent
with principles of TBI/FFI and constructivist pedagogy.
Programs such as OCR arose because of a concern that some students need
explicit instruction in order to learn a given subject matter and do not learn it simply
as a result of exposure through more implicit methods. The rebuttal from
constructivist pedagogy is that not all learning is created equal: some results in long
term retention and some does not. They point to the distinction between assimilation
and accommodation, with the latter resulting in only short term retention of the
96
learning because it isn’t connected and integrated into the students’ existing bank of
knowledge.
TBwI is consistent with an assimilation model of learning. One aspect of this
was the presence of a live interlocutor while revising the rough draft. The student
participants were already familiar with a narrative oral discourse structure between
two live interlocutors. Having received formal English instruction for 3-4 years, the
participants were already familiar with how to orally recount an experience to
another person. Consistent with Vygotsky’s (1986) understanding of written
discourse as a conversation with a blank piece of paper (cognitively more
challenging because of the absence of the reader and a lack of shared contextual
knowledge), they had greater difficulty expressing themselves in written English.
Changing the writing conference with TBwI allowed me to simulate a reader’s
difficulty in understanding the text. Thus the participant was given the opportunity to
compare and contrast existing knowledge (communicating to a live interlocutor with
a shared contextual knowledge) with new knowledge (having a conversation with a
blank piece of paper). The teacher interviews indicated that in at least one case, the
participant had perceived gains in terms of awareness of audience and ability to self-
correct. This was also born out from an analysis of the writing conference transcripts
which had multiple instances wherein the student participant self-corrected a
sentence as s/he read it to me.
Another tenant of constructivist pedagogy is that assimilation can only occur
once a cognitive failure is acknowledged. Given the meaning centered nature of
97
TBwI, the focus was on recasting the participant’s interlanguage into standard
English. The starting point for this process almost invariably began with I don’t
understand. Every time this happened, the student participant was confronted with a
cognitive failure (the student’s interlanguage did not result in an authentic transfer of
meaning from the writer to the reader). The student participant then participated in a
dialogue to recast the sentence into standard English.
Another aspect of constructivist pedagogy was the use of prompts and recasts
as a means of implicit instruction. Clearly, there is already a lot of literature as to
situations where prompts (more explicit) have greater efficacy than recasts (more
implicit); however, within a mixed ability classroom there will be situations where
both will be useful. With TBwI, recasts were used only in situations where the
participants were unable to engage in more active participation in a joint construction
of meaning. It was a tool of last resort. Prompts allowed the participants greater
voice in the process of recasting the draft sentences and as such created more
opportunities for them to actively participate in the learning process.
Within the realm of second language acquisition, there are many and often
contradictory voices. TBwI is consistent with many ideas posited by Krashen and
Terrell’s (2000) natural approach, principally that second language learners need to
have sufficient comprehensible input and that motivation is an important factor.
Providing the second language learner with specific feedback and recasting written
interlanguage into standard English was an effective means of providing
comprehensible input. No prior knowledge nor schema needed to be activated
98
because the writing conference revolved around the participant’s own text-self
connections within their writing. Second, they had a genuine desire to be understood
by me and, thus, there was a strong aspect of motivation, consistent with Krashen’s
affective filter hypothesis (Krashen & Terrell, 2000).
Consistent with Long’s (1996) interaction hypothesis, the participants
(English language learners) were given a chance to dialogue with a native English
speaker in order to make meaning of a given text. Consistent with Swain’s (1985)
output hypothesis, they were required to express themselves in English and take
greater ownership in revising the sentences through a judicious mix of prompts and
recasts.
One of the challenges discussed by Ellis (2006b) and Swan (2005) was the
need for TBI/FFI to be meaning centered and result in an authentic language use
through a communication problem. TBwI satisfied this theoretical ideal of a jigsaw
task because the two interlocutors each had a unique area of expertise. I the native
speaker had linguistic expertise while the participants (non-native speakers) had
experiential expertise. The task was meaning centered and posed a communication
problem that needed to be solved (viz. I don’t understand).
One aspect of constructivism that was not followed were structured
opportunities for prewriting activities. A key tenant to constructivism is that the
student must assimilate new knowledge with the existing knowledge base. This
posed a greater problem with the more cognitively challenging writing prompts (e.g.,
writing about imagination) and certain idioms. For example, Participant H
99
understood the idiom by accident to mean something negative and, thus,
misunderstood the prompt. This could have been better addressed beforehand by
structuring more prewriting activities instead of simply having them write
immediately.
In addition, the process lacked follow-through. I had believed that the
communicative intent would have been to publish (read) these texts to their peers,
and that I would simply be a surrogate interlocutor, doing a task to help them publish
their writing. In fact, I was the intended audience! The participants had a strong
communicative desire for me to understand their ideas. It was this motivation as
much as anything else that sustained them through the hard work of these tasks.
Also noteworthy is that by using prompts (more explicit reactive negative
feedback) and then resorting to recasts (more implicit ) when the prompts did not
work, I was constantly striving to maintain a level of assisted performance attuned to
the their ZPD. In this respect, written tests (measures of independent performance)
are not relevant. In these situations the teacher needs a reliable measure of the
student’s capabilities under conditions of assisted performance. In other words,
TBwI is not a one way flow of information from the teacher to the student. It is a
constant back and forth flow of information; thus, the teacher needs to be as good a
listener as a communicator.
Much of the literature on TBI shows that it has yet to live up to the potential
trumpeted by its proponents. Often the problem is a lack of authentic language use or
authentic communication problem, the students are simply thrown together prompted
100
to talk with each other for lack of native English speakers in the classroom. The tasks
can have an artificial or stilted feel to them. However, making written text-self
connections is an authentic use of language and recasting the written interlanguage
into standard English is in fact an authentic communication problem that needs to be
solved. Thus, TBwI can be effective for pairing up the teacher and student together
in a jigsaw task. Also, pairing up the native and non-native speaker in dyads remains
true to the research of Long (1996), whose participants were paired into dyads of
native and non-native speakers. It is also consistent with the theories of Krashen
(1982) that comprehensible input and motivation (i.e., low affective filter) are key
elements to successful second language acquisition.
Q2, TBwI and Differentiated Instruction
The second research question asked how TBwI impacts the degree of
differentiated instruction within a mixed ability classroom. Clearly, both the
qualitative and quantitative data analysis indicate a high level of differentiation of
instruction through TBwI. Although there was a common element of certain targeted
linguistic features (verb tense usage, complete sentences, writing detail sentences
connected to a give topic sentence), the actual content of each writing conference
varied according to the participant’s instructional needs at the moment because I
responded (provided reactive negative feedback) to the most challenging element of
the written interlanguage within the context of a particular sentence and paragraph.
The mix of explicit/implicit instruction (as measured by the ratio of prompts to
recasts) also varied for each participant. The amount of scaffolding provided (as
101
measured by the ratio of turns to recast sentences) also varied with each participant.
Those who required more support received more while students who needed less
support received less. Thus for these particular students, TBwI was effective in
providing differentiation of instruction.
Q3, TBwI and Second Language Acquisition
The third research question was to understand how TBwI impacts second
language acquisition. This was perhaps the most challenging of questions to answer
because from the data set we must extrapolate to the endpoint of a multiyear process.
The teacher interviews also point to the potential of TBwI as having long term
benefits, at least in the case of some of the participants. The data set does not offer
conclusive information whether there was meaningful long term language
acquisition. It is possible, however, to observe that TBwI represents a teaching
strategy that is consistent with our understanding of the learning process and second
language acquisition theory.
Recommendations for Practitioners
Public education in California, for better or worse, has evolved to have
greater accountability and higher expectations. Public policy makers have sought to
solve societal problems through higher academic expectations. By raising the bar of
higher expectations, it was hoped that student achievement would increase.
This trend, while laudable, has had certain unintended consequences. First,
there has been an increased emphasis on fast-paced direct instruction in classrooms
of low performing students. This grew in part from a belief that low ability students
102
with an external locus of control need a fundamentally different kind of classroom
focused on delivering direct instruction and behaviorist pedagogy.
Rather than wait for such students to grasp a concept through a discovery
approach, it is much faster and efficient (from this perspective) to very explicitly give
the student the answer. In this manner it is possible for a teacher to cover more
material in a shorter period of time. Unfortunately, it has also tipped the balance in
favor of accommodation rather than assimilation (the student actively incorporating
concepts and language into already internalized knowledge).
But students need to be presented with opportunities to test hypotheses under
conditions of assisted performance. The present study placed a strong emphasis on
eliciting language hypotheses from the minority language students through the use of
prompts, provided they were within the participants’ ZPD.
Thus, my first recommendation is that students be given genuine
opportunities to test and reflect upon their language hypotheses and assumptions
about how the second language works. Prompts will not produce novel results if they
are administered by teachers in a formulaic manner that does not allow for this.
The second unintended consequence of raising the bar of academic
expectations has been to equate higher expectations with higher motivation. While
there is a correlation between the two, it is not one-to-one. Human motivation is a
very complex phenomena. Communicative language approaches can be very
effective if they tap into the very powerful motivations for communication: to
understand and be understood, to achieve acceptance and status within a social
103
group. The participants were engaged in the tasks because of a genuine desire to use
language as a social tool. In other words, they were highly motivated.
Thus, my second recommendation is that TBwI be used with writing prompts
that enable authentic written communication, such as tapping into powerful student
experiences that serve as text-self connections. I still remember very clearly the
moment that Participant C recounted her mother asking for her help because she
couldn’t read the note the postman had left. That was a very powerful experience for
her, and she was seemed willing to move heaven and earth to make me understand
that experience as well.
My third recommendation is that TBwI be used as a tool to supplement (not
supplant) other forms of instruction for second language acquisition and writing.
While it is important that students receive some amount of individualized instruction,
it is not feasible to expect that a substantial portion of instruction can be delivered on
such an individualized basis. What is perhaps more reasonable is that they receive
effective instruction in language and concepts through more traditional means, and
that TBwI be simply one additional opportunity for students to test and reflect upon
language hypotheses as well as receive reactive negative feedback and
comprehensible input from a native English speaker (the teacher).
Policy Implications
During the era of whole language, English was taught holistically. Explicit
instruction of text structure, reading strategies, phonics and phonemic awareness was
de-emphasized. During the current era constructivist pedagogy, holistic instruction
104
and a focus on meaning tend to be de-emphasized. Within the arena of second
language acquisition we see a similar phenomenon in terms of the debate regarding
the role of grammar in the instructional syllabus and the optimal point within the
implicit/explicit instructional continuum.
It is clear that completing that volume of writing was no easy task for them.
Listening to someone confront them with their mistakes during the writing
conference was no easy task. Through prompts, I was constantly confronting them
with their own cognitive failures and challenging them to assimilate new
information. A form of learning like this will not work unless the student has a high
degree of motivation.
Therefore, it is noteworthy that this constructivist pedagogy worked for
Participant A (medium), Participant B (high) as well as Participant G (low). They
represent various points along the ability spectrum. What seems to have happened in
California’s classes, according to Achinstein et al. (2004) and Achinstein & Ogawa
(2006) is the development of a two-tier educational system, with students of low
socioeconomic status and minority language students receiving mostly direct
instruction and behaviorist pedagogy. O'Neill (1988) provides the key rationale for
this when he makes an association between low ability pupils and external locus of
control, which is a polite way of saying that they lack motivation to learn. This is
ironic because one of the strongest cultural imperatives of humans is to listen and be
listened to, to use language as an authentic tool for communication. This holds true
for minority language students with mental retardation, those with genius level
105
intelligence, and all points in between. If the teacher is able to tap into this
motivation then they will rise to the challenge of constructivist pedagogy.
Therefore, one of the policy implications of this research is that teachers need
to become more aware of a possible bias towards a two-tier educational system:
education for high ability pupils that is based in constructivist pedagogy and
education for low ability pupils that is based on behaviorist pedagogy and direct
instruction. So called low ability pupils within the classroom may respond well to
more implicit forms of instruction that are centered in constructivist pedagogy,
provided that the student is able to make a connection to existing knowledge and the
teacher is skilled at maintaining instruction within the ZPD.
What can be said with certainty is that not all methods and approaches will
work equally well for all students. Within the context of OCR, this realization has led
to a push for differentiated instruction. Paradoxically this trend has occurred along
with a concurrent push for uniformity of instruction and fidelity to a district intended
curriculum.
Clearly, one of the strongest policy implications that emerges is that research
and theory from second language acquisition and constructivist pedagogy do have a
valid place in the discourse of instruction of minority language students. The analysis
of the data does not indicate OCR and direct instruction to be ineffective nor does it
warrant replacing the current curriculum with 2-1/2 hours a day of mandated TBI,
FFI, or TBwI for the minority language students. However, the literature and this
study do indicate that constructivist pedagogy (to which communicative language
106
teaching is related) does have a valid place with minority language students for at
least some portion of the school day, at least for some students and under some
conditions.
Somehow, educators and education policy-makers need to challenge the
notion that minority language students need direct instruction in place of
constructivist pedagogy because of an association of low ability pupil with external
locus of control. The analysis of the data in this study indicates that, at least for these
third grade minority language students and in this particular context, there was no
correlation between ability level and need for behaviorist pedagogy.
Research Needed
Clearly, more research is needed to follow up this study regarding the
viability of TBwI as an instructional tool. Studies of longer term, with different age
groups and language ability levels, are needed to determine whether TBwI can be an
effective tool for second language acquisition. More research is also needed to
understand the impact of actually incorporating TBwI into the school day of an OCR
classroom instead of simply simulating that through an after school intervention.
Furthermore, more research is needed to understand whether TBwI can be
successfully implemented with students of other ages and stages of language
development.
One of the great challenges of research in the area of second language
acquisition is the long term nature of the process and the multitudinous variables that
affect language acquisition. Clearly it is a complex process; however, it is also a very
107
important process that needs to be better understood so that teacher-practictioners
have a greater understanding of how to connect theory to practice.
Conclusion
Results indicate that TBwI can be a useful tool for providing differentiated
instruction, constructivist pedagogy, and principles of second language acquisition
into the classroom to respond to the diverse needs of the language learners
Upon analysis of the data set and reflection of my personal experience
through this process, one broad theme emerges. To teach with success requires hard
work, content area expertise (in this case, English) and a clear understanding of
teaching pedagogy. But even more important, it requires that the teacher have
empathy for the student. Second language acquisition is far more difficult than
second language teaching. If successful second language teaching can be likened to
climbing Mount Everest, then successful second language acquisition might perhaps
be likened to climbing Jacob’s ladder. Furthermore, learning to have a conversation
with a blank piece of paper is no easy task, neither in a second language nor the first.
An empathetic student-centered approach that harnesses the power of communicative
intent inherent to all languages and a non-dogmatic approach to implementing
teaching pedagogy will not guarantee second language acquisition, but under the
right conditions it will optimize the minority language student’s opportunity to learn.
These right conditions include creating an opportunity where there is a
primacy of meaning that taps into a strong current of motivation. Many times as a
teacher, I have been challenged by a minority language student who was not
108
motivated to complete tasks nor remain focused during a lecture/presentation. The
participants, on the other hand, seemed genuinely interested in helping me to
understand their text-self connections (their written drafts were personal connections
to quotes from literature they were reading during the regular school day). And when
I confronted them with a communication problem (that I didn’t understand one
aspect or another of their writing), they seemed genuinely committed to helping me
to understand their viewpoint. As Smith (1988) pointed out, instructional activities
with no intrinsic meaning that exist only to be graded should be avoided because
they are not effective in tapping into this current of motivation. Bley-Vroman (1988)
noted the distinction between first and second language learning, that the former is
almost universally successful while the latter frequently is not. Perhaps one reason is
the very strong motivation (often a matter of survival) of the infant to understand and
be understood by his/her caretakers. Using TBwI will not guarantee a high degree of
motivation and success, but at least within the context of this study it did effectively
maintain student interest and motivation to have a conversation with a blank piece of
paper.
109
GLOSSARY
Assimilation: In constructivist pedagogy, the process by which new information is
incorporated with knowledge already internalized (De Lisi & Golbeck, 1999)
Behaviorist Pedagogy: A teaching method that emphasizes the ability to learn new
information through repetition and structured practice.
Coherency: The degree to which sentences within a paragraph are connected to each
other through a given topic sentence (Celce-Murcia & Olshtain, 2005).
Constructivist Pedagogy: A type of teaching that emphasizes the need for a learner to
actively participate in the learning process by incorporating new information
into his/her previously internalized knowledge base (De Lisi & Golbeck,
1999).
Cross-Cultural Language and Academic Development (CLAD): A supplementary
certificate for teaching credentials issued by the state of California
authorizing teachers to provide English instruction to students whose first
language is not English.
Differentiated Instruction: Instruction that varies according to the instructional needs
of students in a multi-level or multi-ability classroom.
Direct Instruction Pedagogy: A type of teaching method that emphasizes fast-paced
and explicit instruction that may be highly scripted O'Neill (1988).
English Immersion Instruction: A instruction for public school students whereby the
teacher may not speak in the students’ first language except to clarify
directions.
English Language Learner: A student who is learning English as a second language.
Fabric of Concepts: Vygotsky (1986) used this metaphor to the characterize the
interconnected nature of a child’s internalized bank of knowledge.
Focus on Form: Long & Robinson (1998) describes this as a type of language
instruction that is holistic, focused on meaning, with some grammar
instruction within the context of a focus on meaning.
Focus on FormS: Long & Robinson (1998) describes this as a type of language
instruction that is a grammatical syllabus focused on specific grammar topics.
110
Focused Tasks: (Ellis, 2003) described it as type of second language instruction that
is designed by the teacher to create opportunities for instruction to focus on a
specific grammar topic.
Form Focused Instruction (FFI): An implicit form of grammar instruction, within
the umbrella of communicative language teaching, and associated with task
based instruction, and may incorporate focused tasks (Skehan, 1998).
Independent Work Time: Bereiter et al. (2000) describes it as a 20-40 minute period
of instruction within OCR when most of the students are engaged in
independent activities, giving the teacher the opportunity to work provide
differentiated instruction by working one-on-one or with flexible small
groups of students to pre-teach or re-teach OCR lessons.
Information-Gap Task: Ellis (2003) describes it as a a type of task based instruction
with two participants, one who holds the information and the other who does
not. To complete the task this information must be exchanged.
Input Flood: Ellis (2003) describes it as an implicit form of grammar instruction
whereby students are exposed to written or oral language that contains a high
concentration of a particular linguistic form, thereby encouraging students to
become aware of it while still remaining focused on meaning.
Interlanguage: the second language learner’s emerging use of the second language
that varies from a native speaker’s use of the second language.
Interaction Hypothesis: Long (1996) posited that language acquisition that is both
expressive and receptive requires more than comprehensible input, it also
requires sufficient opportunities for conversational interaction.
Interlocutor: a person engaged in the exchange of information through oral or
written language.
Jigsaw Task: Ellis (2003) describes it as a a type of two-way information-gap task
(task based instruction). The two participants each hold information needed
to complete the task, and an exchange of that information is needed to
complete the task.
Minority Language Student: within the United States, one who is learning English as
a second language; this term is similar to English Language Learner except
that it emphasizes that the students primary language is not that of the
dominant culture.
111
Negotiation of Form: the process that two speakers use to call attention to issues of
linguistic (grammar) issues rather that a strict focus on meaning. Often there
is a contrast between the interlanguage of one of the interlocutors and
standard usage of the second language.
Open Court Reading: An English language arts textbook program published by
McGraw Hill for use in kindergarten through sixth grade classrooms.
Program Improvement: Consistent with federal legislation termed No Child Left
Behind, a designation for schools that fail to meet annual federal targets for
academic improvement as measured by state-wide standardized tests.
Prompt, Clarification: Lyster (2004) defined it as a conversational technique teachers
can use (more explicit than recasts) to help the second language learner
understand that what s/he said is misunderstand or has linguistic errors.
Prompt, Elicitation: Lyster (2004) defined it as a conversational technique teachers
can use (more explicit than recasts) to encourage the second language learner
to complete the end of the teacher’s utterance.
Prompt, Metacognitive: Lyster (2004) defined it as a conversational technique
teachers can use (more explicit than recasts) to encourage the second
language learner to become aware of his/her thinking and/or grammar usage.
Reading Fluency: A measure of speed and accuracy reading aloud within OCR. The
student reads an unfamiliar passage for one minute and is the count of correct
words read (total words correctly read aloud minus mistakes).
Recast: When a second language learner says something with linguistic errors, the
teacher may rephrase (hence recast) it, used as an implicit form of instruction
to help the language learner notice differences between his/her interlanguage
and standard usage of the second language (Elis, 2003).
Schema: a term for the collective bank of knowledge a person has, emphasizing the
interconnectedness of internalized knowledge (Celce-Murcia & Olshtain,
2000).
Semantics: how meaning is encoded within words
Sentence Lifting: A process within OCR whereby a teacher takes a short writing
sample of a student and revises it during whole group instruction.
112
Specially Designed Academic Instruction in English (SDAIE): A pedagogy of second
language instruction that uses various techniques to support the learning of
concepts and knowledge in a second language (Genzuk, 2003).
Task Based Instruction (TBI): A communicative language teaching approach
whereby language learners must work together and use the second language
to solve an authentic communicative problem (Ellis, 2003)
Task Based writing Instruction (TBwI): A type of task based instruction whereby an
oral language exchange between the teacher and second language learner is
used to rephrase written interlanguage into standard English. The oral
language exchange to rewrite the student’s rough draft maintains primacy of
meaning with an occasional shift in focus to grammar instruction.
Teaching English to Speakers of Other Languages (TESOL): A course of study for
language teachers, focused on pedagogy and theory from second language
acquisition.
Reading Wars, The: A struggle within education and public policy in the United
States (from the 1980s through the present), primarily between proponents of
whole language on one hand and those advocating explicit instruction of
phonics and phonemic awareness on the other (Pearson, 2004).
Topicality: Similar to the concept of cohesion, it is the degree to which supporting
sentences within a paragraph are able to develop a given topic sentence.
Unfocused Tasks: Activities within task based instruction that are not purposefully
designed by the teacher to elicit a particular linguistic form (Ellis, 2003).
Whole Language: A type of English language pedagogy in which literacy is taught
holistically (to varying degrees); specific literacy skill development may be
embedded within reading and writing activities rather than through explicit
instruction, similar to communicative language approaches but originally
intended for native-English speakers.
Zone of Proximal Development: Represents skills and concepts maturating within a
learner, which s/he may use to solve problems under conditions of assisted
performance; Vygotsky posited that two students may have the same level of
independent performance but have very different instructional needs because
one may have a larger zone of proximal development. It is an important
theoretical rationale for differentiated instruction.
113
REFERENCES
Achinstein, B. & Ogawa, R. T. (2006). (In)Fidelity: What the resistance of new
teachers reveals about professional principles and prescriptive educational
policies. Harvard Educational Review, 76(1), 30-63.
Achinstein, B., Ogawa, R. T., & Speiglman, A. (2004). Are we creating separate and
unequal tracks of teachers? The impact of state policy, local conditions, and
teacher characteristics on new teacher socialization. American Educational
Researcher Journal, 41(3), 557-603.
Ammar, A. & Spada, N. (2006). One size fits all? Recasts, prompts, and L2 learning.
Studies in Second Language Acquisition, 28(4), 543-574.
Barton, P. E. (2005). One-third of a nation: Rising dropout rates and declining
opportunities. Retrieved April 24, 2006, from http://www.ets.org/research/
researcher/pic-onethird.html
Beck, I. L., McKeown, M. G., & Kucan, L. (2002). Bringing words to life: Robust
vocabulary instruction. New York: Guilford Press.
Bereiter, C., Brown, A., Campione, J., Carruthers, I., Case, R., Hirshberg, J., et al.
(2000). Open court reading (teacher's edition), level 5, book 1. Worthington,
Ohio: SRA/McGraw-Hill.
Beveridge, M. (1997). Educational implementation and teaching. In L. Smith, J.
Dockrell, & P. Tomlinson (Eds.), Piaget, Vygotsky and beyond. (pp. 21-35).
New York, NY: Routledge.
Cazden, C. B. (2001). Classroom discourse: The language of teaching and learning.
Portsmouth, NH: Heinemann.
Celce-Murcia, M. & Olshtain, E. (2000). Discourse and context in language
teaching. New York: Cambridge University Press.
Celce-Murcia, M. & Olshtain, E. (2005). Discourse-based approaches: A new
framework for second language teaching and learning. In E. Hinkel (Ed.),
Handbook of Research in Second Language Teaching. (pp. 729-741).
Mahwah, NJ: Lawrence Erlbaum.
Conner, C. M. D., Morrison, F. J., & Katch, L. E. (2004). Beyond the reading wars:
Exploring the effect of child-instruction interactions on growth in early
reading. Scientific Studies of Reading, 8(4), 305-336.
114
Crawford, J. (2004). Educating English learners: Language diversity in the
classroom. Los Angeles: Bilingual Education Services, Inc.
Cummins, J. (1992). Language proficiency, bilingualism, and academic achievement.
In P. Richard-Amato & M. A. Snow (Eds.), The multicultural classroom:
Readings for content-area teachers. (pp. 16-26). Reading, MA: Addison
Wesley Publishing Company.
Cummins, J. (1994). Primary language instruction and the education of language
minority students. In C. F. Leyba (Ed.), Schooling and language minority
students: A theoretical framework. (pp. 3-46). Los Angeles: Legal Books
Distributing.
De Lisi, R. & Golbeck, S. L. (1999). Implications of Piagetian theory for peer
learning. In A. O'Donnell & A. King (Eds.), Cognitive perspectives on peer
learning. (pp. 3-37). Mahwah, NJ: Lawrence Erlbaum Associates.
Doughty, C. (2001). Cognitive underpinnings of focus on form. In P. Robinson (Ed.),
Cognition and second language instruction. New York: Cambridge
University Press.
Doughty, C. & Varela, E. (1998). Communicative focus on form. In C. Doughty & J.
Williams (Eds.), Focus on form in classroom second language acquisition.
(pp. 114-138). New York: Cambridge University Press.
Ellis, D. R. (2003). Task-based language learning and teaching. New York: Oxford
University Press.
Ellis, R. (2006a). Modeling learning difficulty and second language proficiency: The
differential contributions of implicit and explicit knowledge. Applied
Linguistics, 27(3), 431-463.
Ellis, R. (2006b). Reexamining the role of recasts in second language acquisition.
Studies in Second Language Acquisition, 28(4), 575-600.
Ferris, D. (2002). Treatment of error in second language student writing. Ann Arbor,
MI: University of Michigan Press.
Foorman, B., Francis, D. J., Fletcher, J. M., Mehta, P., & Schatschneider, C. (1998).
The role of instruction in learning to read: Preventing reading failure in at-
risk children. Journal of Educational Psychology, 90(1), 37-55.
115
Gallego Codes, J. (2004). Enseñar a pensar en la escuela. [Teaching to think in the
school] Madrid, Spain: Ediciones Pirámide.
Genzuk, M. (2003). Specially designed academic instruction in English (SADIE) for
language minority students. Unpublished manuscript.
Gibbons, P. (1998). Classroom talk and the learning of new registers in a second
language. Language and Education, 12(2), 99-118.
Grice, H. P. (1975). Logic and conversation. In P. Cole & J. Morgan (Eds.), Syntax
and semantics, vol. 3, speech acts. (pp. 41-58). New York: Academic Press.
Haneda, M. (2004). The joint construction of meaning in writing conferences.
Applied Linguistics, 25(2), 178-219.
Harley, B. (1998). The role of focus-on-form tasks in promoting child L2 acquisition.
In C. Doughty & J. Williams (Eds.), Focus on form in classroom second
language acquisition. (pp. 156-174). New York: Cambridge University Press.
Izumi, S., Bigelow, M., Fujiwara, M., & Fearnow, S. (1999). Effects of output on
noticing and second language acquisition. Studies in Second Language
Acquisition, 21, 421-452.
Izumi, L., Coburn, K. G., & Cox, M. (2002). They have overcome: High-poverty,
high-performing schools in California. Retrieved September 9, 2006, from
www.pacificresearch.org
Krashen, S. D. & Biber, D. (1988). On Course: Bilingual education's success in
California. Ontario, CA: California Association for Bilingual Education.
Krashen, S. D. (1982). Principles and practice in second language acquisition. New
York: Pergamon.
Krashen, S. D. & Terrell, T. D. (2000). The natural approach: Language acquisition
in the classroom. Harlow, Essex: Pearson Education Limited.
Landsberg, M. (2006, March 2). L.A. mayor sees dropout rate as 'civil rights' issue.
Los Angeles Times, pp. B1. Retrieved November 9, 2006 from http:/
/www.latimes.com/news/local/la- me-
dropout2mar02,1,5830485.story?ctrack=1&cset=true.
116
Lightbown, P. M. (1998). The importance of timing in focus on form. In C. Doughty
& J. Williams (Eds.), Focus on form in classroom second language
acquisition. (pp. 177-196). New York: Cambridge University Press.
Long, M. H. (1996). The role of the linguistic environment in second language
acquisition. In W. Ritchie & T. Bhatia (Eds.), Handbook of second language
acquisition. (pp. 413-468). New York: Academic Press.
Long, M. H., Inagaki, S., & Ortega, L. (1998). The role of implicit negative feedback
in SLA: Models and recasts in Japanese and Spanish. Modern Language
Journal, 82(3), 357-371.
Long, M. & Robinson, P. (1998). Focus on form: Theory, research, and practice. In
C. Doughty & J. Williams (Eds.), Focus on form in classroom second
language acquisition. (pp. 15-41). New York: Cambridge University Press.
Losen, D. & Wald, J. (2005). Confronting the graduation rate crisis in California.
Retrieved April 24, 2006, from http://www.civilrightsproject.harvard.edu/
research/dropouts/March%2023%20California%20Final%20Report.pdf
Lyster, R. (2004). Differential effects of prompts and recasts in form-focused
instruction. Studies in Second Language Acquisition, 26, 399-432.
Mackey, A. (2006). Feedback, noticing and instructed second language learning.
Applied Linguistics, 27(3), 405-443.
Maddahian, E. (2002). A comparative study of second grade students' reading,
language, and spelling gain scores for the Los Angeles Unified School
District reading programs. Retrieved September 9, 2006, www.eric.gov/
Manzo, K. K. (2004, September 15). L.A. students get reading by the book.
Education Week, 24(3), 16-18.
Morey, N. C. & Luthans, F. (1984). An emic perspective and ethnoscience methods
for organizational research. The Academy of Management Review, 9(1),
27-36.
Moustafa, M. & Land, R. (2001, November). The effectiveness of "Open Court" on
improving the reading achievement of economically-disadvantaged children.
Paper presented at the Annual Meeting of the National Council of Teachers
of English, Baltimore, MD.
117
National Institute of Child Health and Human Development. (2000). Report of the
National Reading Panel. Teaching children to read: An evidence-based
assessment of the scientific research literature on reading and its
implications for reading instruction (NIH Publication No. 00-4769).
Washington, DC: U.S. Government Printing Office.
Norris, J. & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis
and quantitative meta-analysis. Language Learning, 50(3), 417-528.
O'Neill, G. P. (1988). Teaching effectiveness: A review of the research. Canadian
Journal of Education, 13(1), 162-185.
Pearson, P. D. (2004). The reading wars. Educational Policy, 18(1), 216-252.
Peck, S. & Serrano, A. M. (2002, April). "Open Court" and English language
learners: Questions and strategies. Annual Meeting of the American
Association for Applied Linguistics Retrieved September 6, 2006, from
www.eric.gov/
Pica, T. (2005). Classroom learning, teaching and research: A task-based perspective.
The Modern Language Journal, 89(3), 339-352.
Reading/language arts framework. (1999). Sacramento, CA: CDE Press.
Resnick, L. & Nelson-LeGall, S. (1997). Socializing intelligence. Piaget, Vygotsky
and beyond. (pp. 145-158). London: Routledge.
Richards, J. C., Rodgers, T. S., & Swan, M. (2001). Approaches and methods in
language teaching. Cambridge University Press.
Rogers, J., Terriquez, V., Valladares, S., & Oakes, J. (2006). California educational
opportunity report 2006. Retrieved November 10, 2006, from http:/
/www.idea.gseis.ucla.edu/publications/eor06/fullreport/index.html
Rumberger, R. W., Gándara, P., & Merino, B. (2006). Where California's English
learners attend school and why it matters. UC LMRI newsletter, 15(2), 1-3.
Schmidt, R. (2001). Attention. In P. Robinson (Ed.), Cognition and second language
instruction. New York: Cambridge University Press.
Skehan, P. (1998). A cognitive approach to language learning. New York: Oxford
University Press.
118
Smith, F. (1988). Joining the literacy club. Portsmouth, NH: Heinemann.
Swain, M. (1985). Communicative competence: Some roles of comprehensible input
and comprehensible output in its development. In S. Gass & C. Madden
(Eds.), Input in second langauge acquisition. (pp. 235-253). Rowley, MA:
Newbury House.
Swan, M. (2005). Legislation by hypothesis: The case of task-based instruction.
Applied Linguistics, 26(3), 376-401.
Thomas, W. P. & Collier, V. P. (2002). A national study of school effectiveness for
language minority students' long-term academic achievement final report:
Project 1.1. Retrieved November 12, 2006, from http://www.crede.org/
research/llaa/1.1_final.html
Vygotsky, L. (1978). Mind in society. Cambridge, MA: Harvard University Press.
Vygotsky, L. (1986). Thought and language. Cambridge, MA: The MIT Press.
Warren, P. (2004). A look at the progress of English language learner students.
Retrieved July 28, 2005, from http://www.lao.ca.gov/2004/english_learners/
021204_english_learners.pdf
Williams, T., Kirst, M., Haertel, E., Woody, E., Levin, J., Studier, C. et al. (2005).
Similar students, different results: Why do some schools do better? Retrieved
September 9, 2006, from http://www.edsource.org/pdf/SimStu05.pdf
Wilson, P. G., Martens, P., Arya, P., & Altwerger, B. (2004). Readers, instruction,
and the NRP. Phi Delta Kappan, 86(3), 242-246.
119
LIST OF APPENDICES
Appendix A: Participant Summary Data 130
Appendix B: Writing Prompts 131
Appendix C: Sample Writing Conference 133
Appendix D: Recruitment Tools 137
Appendix E: Development of Criteria for Data Coding 140
120
APPENDIX A: SUBJECT SUMMARY DATA
Participant A
Participant A participated in three writing conferences, during weeks 1, 2,
and 4 of the intervention, with a total of 18 sentences and 50 turns. The primary
focus of the writing conferences was verb tense usage. Each of the three writing
prompts he chose focused on a different verb tense (present, past, and future). There
were many issues that I could have pursued within the reactive instruction, but I
chose to focus mostly on verb tense usage. He averaged 2.78 turns per recast
sentence.
Figure 6: Participant A Linguistic Focus
Verb Tense
52%
Schema
16%
Simile
16%
Syntax
7%
End Marks
5%
Topicality
3%
Verb (Other)
1%
Other
8%
The time spent on the simile category represents a single exchange in a single
conference, nine turns to recast It was like brand new to be It was like a brand new
121
toy. This was extremely challenging for him, but is an excellent example of
Vygotsky’s concept of assisted performance; under conditions of assisted
performance (guided questions), he was able to create a sentence of figurative
language, something he could not have done independently.
Participant B
Participant B participated in 7 writing conferences, by far the most prolific
writer of the group, with 40 recast sentences over 94 turns. He averaged 2.35 turns
per recast sentence.
Figure 7: Participant B Linguistic Focus
Schema
50%
Coherency
18%
Topicality
11%
Pronouns
6%
Quotations
5%
Simile
3%
Prepositions
2%
Verb Tense
2%
Syntax
2%
End Marks
1%
Other
7%
He had very few problems with verb tense or complete sentences. So together
we focused on issues of overall structure and meaning rather than specific linguistic
features. Problems with schema often took the form of narratives where there was
122
not sufficient information. For example, in one writing conference he wrote about
fixing a problem dealing with a specific game, but without knowledge of the game’s
rules it made no sense. By adding additional information to clarify this point, the
paragraph became meaningful. In another instance, he wrote about using his
imagination during play, but spent much of that draft discussing what happened at
home later that day after the imaginative play episode was concluded. Teacher
interview indicated that “he became better at editing his own work.”
Participant C
Participant C had one conference during week 1 and two during week 3. She
had 17 sentences recast over 42 turns, with an overall ratio of 2.47 turns per recast
sentence.
Figure 8: Participant C Linguistic Focus
Schema
22%
Semantics
19%
Quotations
18%
End Marks
12% Prepositions
7%
Pronouns
6%
Syntax
6%
Verb (Other)
6%
Coherency
4%
Other
15%
123
As is reflected above, she had very balanced needs, requiring support in
multiple areas to encode her thoughts into written standard English.
In one very important conference, she wrote about how she had helped her
mother once. From her writing I could not understand what kind of support she had
given (nor would other readers have understood with a shared schema). Only through
a series of guided questions was I able to understand that her mother could not read
the note the postman had handed her. It was a very powerful moment for both of us.
Teacher interview indicated that “she used to write with lots of run-on sentences, and
sometimes her sentences would make no sense, or she wasn’t producing what she
was trying to communicate.”
Participant D
Participant D had five conferences. He had 26 sentences recast over 94 turns,
Figure 9: Participant D Linguistic Focus
End Marks
30%
Schema
26%
Topicality
11%
Quotations
9%
Verb Tense
8%
Coherency
7% Verb (Other)
3%
Syntax
3%
Prepositions
2%
Commas
1%
Other
9%
124
with an overall ratio of 3.62 turns per recast sentence. He had very little problem
with verb tense usage, which was the initial focus of the investigation. He had fairly
good writing fluency, filling between 1 and 1-1/2 pages for each draft he completed.
The area of greatest need was the use of end marks and learning the concept of what
exactly is a complete sentence. He was a very quiet student during the interactions,
often giving one-word responses to my queries during TBwI, thus I found myself
adjusting to use more closed ended questions because he would be silent with open
ended ones.
He demonstrated growth over time as evidenced by his last rough draft,
which showed correct use of end marks and the punctuation rules for quotes (comma
& quote marks). Teacher interview indicated that he “narrowed down that thinking
time and actually produced. He would be able to pair/share with somebody else and
... then get right to it.”
Participant E
Participant E only participated in three writing conferences, with 12
sentences recast over 36 turns, averaging three turns per recast sentence.
The data on the distribution focus reflects that each writing conference
focused on a very specific area of need. For example, the second writing conference
dealt with the use of imagination, something he didn’t understand well.
The amount of time spent on Topicality reflected that he was writing about
something he really didn’t understand. The quantity of time spent on Schema implied
that it was not an important aspect, although it really was. He wrote one whole paper
125
Figure 10: Participant E Linguistic Focus
End Marks
25%
Topicality
26%
Verb Tense
21%
Commas
9%
Coherency
9%
Schema
5%
Semantics
5%
Other
18%
about finding something valuable once, but through his draft he never actually said
what it was (a gold ring).
Participant F
Participant F only participated in two conferences during the last week of the
study, with 17 sentences recast over 46 turns, averaging 2.17 turns per recast
sentence.
The greatest area of difficulty for her was Topicality. Similar to Participant
E, she wrote about imagination, a topic she did not really understand. What is
interesting is that through assisted performance she was able to demonstrate
understanding of the concept by self-correcting a sentence with faulty grammar
during the writing conference (reading I drawed a frog as I imagined a frog
dancing).
126
Figure 11: Participant F Linguistic Focus
Topicality
40%
Semantics
16%
Schema
10%
Pronouns
10%
End Marks
8%
Quotations
7%
Coherency
5%
Verb Tense
2%
Prepositions
2%
Other
9%
The other paragraph dealt with a scary moment and she had included
insufficient information for the reader to understand what exactly was scary about
that event.
Participant G
Participant G was by far the lowest student in terms of language
development and literacy skills. She participated in five writing conferences but only
in three cases were they officially included because she was not able to read her own
writing (nor was I able to). She had 19 sentences recast over 76 turns, averaging 4
turns per recast sentence.
Once I attempted to have her orally dictate the paragraph to me but I had to
abandon that idea because she spoke so slowly. However, after a couple of tries she
was determined to create a written piece that would actually contain meaning. She
127
recruited another participant to help her write it and practiced reading her own work.
Thus, was she able to successfully complete three writing conferences. She also had
a very low level of language development and even an expression such as hard times
in the writing prompt was very problematic for her.
Figure 12: Participant G Linguistic Focus
Verb Tense
56%
Quotations
17%
Schema
15%
End Marks
8%
Syntax
3%
Coherency
1%
Other
4%
Participant H
Participant H participated in three conferences, one in the first week and two
in the last week, having 14 recast sentences over 56 turns, with a ratio of four turns
per recast sentence.
Like the other participants, one area of weakness was understanding the
difficulties a reader lacking a common schema would have. (e.g., use of pronouns).
He wrote one paper that included we fell down but never made clear who the we was
128
(he also used we to express actions which obviously were done by just one person).
He also demonstrated difficulty with certain idiomatic expressions, converting the
prompt sometimes you could find something by accident to mean something bad.
Thus we ended up changing the prompt to adjust to the topic he actually wrote about:
sometimes you could break something by accident.
Figure 13: Participant H Linguistic Focus
Pronouns
21%
Topicality
20%
End Marks
21%
Verb Tense
18%
Schema
11%
Semantics
4%
Verb Tense
3%
Prepositions
2%
Other
8%
Participant I
Participant I participated in four conferences, with 29 sentences recast over
149 turns, averaging 5.13 turns per recast sentence.
She had the greatest difficulty as a meaningful participant in the task. She had
very high writing fluency but was resistant to implicit forms of instruction. For
example, when given a prompt to write about a future event (when I grow up, I will
teach my kids to listen and pay attention), she wrote it in the past tense because she
129
was actually writing about a dream she had about what would happen when she
would grow up someday.
Figure 14: Participant I Linguistic Focus
Verb Tense
37%
Topicality
30%
Schema
13%
Pronoun
9%
End Marks
6%
Quotations
4%
Prepositions
1%
Other
11%
130
APPENDIX B: WRITING PROMPTS
IWT WRITING
Tuesday, January 16, 2007
Prompt 1:
“So some days became treasure-hunting days, with everybody trying to find that special kind.
And then on other days you might just find one without even looking.”
(Roxaboxen, p. 260)
I remember a time I found something special.
[add 4 detail sentences]
OR
Sometimes you can find something special by accident.
[add 4 detail sentences]
OR
When I get older, I want to go treasure-hunting.
[add 4 detail sentences]
Prompt 2:
“All you needed for a horse was a stick and some kind of bridle, and you could gallop any-
where.” (Roxaboxen, p. 262)
I remember a time I used my imagination to play.
[add 4 detail sentences + 1 concluding sentence]
OR
If you use your imagination, you can play without a lot of stuff.
[add 4 detail sentences + 1 concluding sentence]
OR
When I grow up, I will teach my kids to use their imagination.
[add 4 detail sentences + 1 concluding sentence]
! Page 1 of 1
/Users/mercuris/Documents/.../Prompt20070116.pages
131
IWT WRITING
Monday, January 29, 2007
Prompt 1:
“But when Jacob was eleven and Wilhelm was ten, their father died. Being the oldest, the
two boys helped hold the family together during this difficult time for their mother.”
(A Tale of Brothers Grimm, p. 24)
I remember a time I helped my family.
[add 4 detail sentences]
OR
It is important to help your parents in hard times.
[add 4 detail sentences]
OR
When I grow up, I will teach my kids help me.
[add 4 detail sentences]
Prompt 2:
“One day they were lucky enough to meet what they called a ‘genuine storyteller.’ ... She
was a widow who sold eggs and butter in the small village where she lived with her children and
grandchildren.” (A Tale of Brothers Grimm, p. 31)
I remember a time I was really lucky.
[add 4 detail sentences + 1 concluding sentence]
OR
With hard work and good luck, you can do anything.
[add 4 detail sentences + 1 concluding sentence]
OR
When I grow up, I will work hard and maybe be lucky too.
[add 4 detail sentences + 1 concluding sentence]
! Page 1 of 1
/Users/mercuris/Documents/.../Prompt20070129.pages
132
APPENDIX C: SAMPLE WRITING CONFERENCE
January 19, 2007

I remember a time I tried
something new and different. I
was eight years old. My dad
and I went to the building. I was
tired. My dad said, “There’s dif-
ferent stuff in the building.
You will have fun with new
friends.” But I wanted to keep
my old friends instead of making
new ones.
I remember a time I tried something new and different.
I was eight years old.
Me and my dad went to the building.
My dad said there’s differenet sufft in the building.
You will have fun with new friends.
But I like my old friends.
! Page 1 of 1
/Users/mercuris/Documents/.../c6.pages
133
page 1 of 3
1
737
738
739
740
741
742
743
744
745
746
747
748
749
A C D E F G H I J M N O Q R
Num Row ID T S Sentence Recast Teacher Student LenT LenS RatioTS P Target
c06 101 I T1 S1
I remember a time I tried
something new and
different.
I remember a time I tried
something new and
different.
okay, why don’t you read me the
first sentence.
okay, I’m ready ... I remember a
time I tried something new and
different.
47 74 0.64 Rea n/a
c06 102 I T2
okay, wait wait ... “Something new
and different” ... So what I’m going
to do is ... you’re going to read me
each sentence ... we’re going to
stop and I’m going to see if I have
any questions or if I have trouble
understanding ... and then we’ll
move on to the next one
- 269 1 269 Sta n/a
c06 103 I T3 S2 I was eight years old. I was eight years old.
okay, so what is the next
sentence?
I was eight years old ... Me and
my dad
35 39 0.9 Rea n/a
c06 104 I T4
Okay, “I was eight years old” ...
that’s really like one sentence,
right? ... okay, so let’s call that a
sentence
- 113 1 113 Sta M.End
c06 106 I T6
Okay so that’s the next sentence,
right?
yes 40 3 13.33 Rea n/a
c06 109 I T9
very good, so I ... so we can
change this to be “My dad and I
went to the building”
- 83 1 83 Sta Pronoun
c06 110 I T10 S4 I was tired. okay, so what was the next one? I was tired ... 31 15 2.07 Rea n/a
c06 112 I T12
okay, so when you are going
through here, one of the things
you are going to check is I want
you to notice where I am putting
the periods, and you are going to
put them the same way
- 181 1 181 Sta M.End
c06 113 I T13 S5
My dad said there’s
differenet sufft in the
building.
My dad said, “There’s
different stuff in the building.
okay so what is the next one?
My dad said there is different stuff
in the building ... you will have fun
with your friends.
29 93 0.31 Rea n/a
c06 116 I T16
marks ... okay, “My dad said
There’s ... okay so comma,
quotation marks, capital ‘T’ there’s
different stuff in the building
- 124 1 124 Sta M.Quote
c06 117 I T17 S6
You will have fun with new
friends.
You will have fun with new
friends.”
okay, then what is the next
sentence?
You will have fun with your friends
but I like my old friends ...
37 65 0.57 Rea n/a
c06 119 I T19
Okay, so we are going to put the
quotation marks, so the quotation
marks are going to end there.
- 96 1 96 Sta M.Quote
c06 120 I T20 S7 But I like my old friends.
But I wanted to keep my old
friends instead of making
new ones.
and then, what is the last
sentence?
but I like my old friends. 36 26 1.38 Rea n/a
12/18/07: 3:44 PM datos.xls: Turn_Detail amb
134
page 2 of 3
1
A C D E F G H I J M N O Q R
Num Row ID T S Sentence Recast Teacher Student LenT LenS RatioTS P Target
787
788
789
790
791
792
793
794
795
796
797
c06 105 I T5 S3
Me and my dad went to the
building.
My dad and I went to the
building.
okay, what does the next
sentence say?
“Me and my dad went to the
building” ... it was
38 47 0.81 Met Pronoun
c06 107 I T7
So, “Me and my dad went to the
building.” ... Just one thing, I
notice here ... Me go to school,
does that sound right? ... Me go to
no 139 2 69.5 Met Pronoun
c06 108 I T8
It shouldn’t be “me went to
school,” it should be ...
I went to school 53 16 3.31 Eli Pronoun
c06 111 I T11
and that is one sentence, right? ...
now, how many periods do we
see here? ... one, two ... is this
two sentences?
no 114 2 57 Met M.End
c06 114 I T14
Okay, wait ... “In the building...”
okay, one question here ... are we
actually using Dad’s real words ...
are we using Dad’s exact words?
yeah 138 4 34.5 Cla M.Quote
c06 115 I T15
Okay, so then we want to put our
... Quo
tation 40 6 6.67 Eli M.Quote
c06 118 I T18
so, let’s stick with this for right
now. “You will have fun with new
friends” ... now, is this still your
dad talking?
yes 118 3 39.33 Cla M.Quote
c06 121 I T21 but I like my old friends too? yes 30 3 10 Rep Schema
c06 122 I T22
I don’t understand, are you saying
that you want to keep your old
friends? or you want to make new
friends but you also want to keep
your old friends? ... I don’t
understand.
It’s because, I don’t like to be
somebody’s friends ... I only like
my old friends.
175 83 2.11 Cla Schema
c06 123 I T23
So you didn’t want to have new
friends, you just wanted to keep
your old friends, right? ... Okay so
let’s say that because if I say “I
like my old friends” it doesn’t really
tell me what you want, right?
no 204 2 102 Met Schema
c06 124 I T24
Okay, so let’s change “but I want,
but I wanted ... what?
I wanted my new friends 57 23 2.48 Cla Schema
12/18/07: 3:44 PM datos.xls: Turn_Detail amb
135
page 3 of 3
1
A C D E F G H I J M N O Q R
Num Row ID T S Sentence Recast Teacher Student LenT LenS RatioTS P Target
798
799
800
c06 125 I T25
You wanted your new friends? or
you wanted to keep your old
friends ... I wanted to keep my
OLD friends. ... and you wanted to
keep your old friends and make
new ones, or you wanted old
friends INSTEAD of new ones?
When I was watching the stairs, I
sort of think about something ...
My dad opened the door, and we
went to the room 27 ... it was like,
third grade, and he let me stay
there for a week, then my dad
went to his work.
214 215 1 Cla Schema
c06 126 I T26
Okay, so did you want to keep ...
So you didn’t really want to make
new friends, right?
no 87 2 43.5 Cla Schema
c06 127 I T27
Okay, so we are going to say, “I
wanted to keep my old friends
INSTEAD of making new ones ...
Okay?
yes 99 3 33 Cla Schema
12/18/07: 3:45 PM datos.xls: Turn_Detail amb
136
APPENDIX D: RECRUITMENT TOOLS
Page 1 of 1
Macintosh HD:Users:mercuris:Documents:Personal:Education:UscEduc594:Admin:IntroLetter2.doc amb
University of Southern California
Rossier School of Education
Waite Phillips Hall, 1001D
Los Angeles, CA 90089-4031
Volunteers Needed for a Writing Study
Hace falta voluntarios para una investigación sobre escritura
Hello, my name is Alex Bantis. I’m a teacher here at xxxxx School. I am also a
student. I go to school at USC. I’m getting a masters degree about teaching English. Just
like your kids have to do research projects, I have to also. I want to do a research project
about learning English and learning to do writing.
I am going to do a research project after school for about a month. It will be for 8-10
kids, from 3:00 until 3:45 in my classroom (room 9). I will give the kids snacks, and as a
thank you for participating in the study they will receive a fountain pen and bottle of ink
(about $35). I am looking for third graders who are learning English as their second
language.
I can’t promise that your child will improve their writing, and remember that it won’t
change what kind of grade your teacher will give your child. Also, they can quit at any time.
If you have any questions, you can contact me directly at (xxx)xxx-xxxx.
Hola, mi nombre es Alex Bantis. Soy un maestro aquí en la escuela xxxxx. También
soy estudiante en USC. Estoy estudiando para una maestría sobre enseñanza de inglés.
Como sus hijos deben hacer projectos y tareas, yo también. Quiero hacer una investigación
sobre aprendiendo inglés, especialmente en escritura.
Voy a hacer una investigación despues del día escolar para un mes. Habrá lugar
para 8-10 estudiantes, desde las 3:00 p.m. hasta 3:45 p.m. en mi aula (salon 9). Los
estudiantes van a recibir un poco de comida para picar. Para dar gracias por auydarme en
mi investigación, su hijo recibirá una pluma fuente y fuente de tinta (valen como $35). Busco
estudiantes del tercer grado que aprenden inglés como idioma segundo.
No puedo prometerle a usted que su hijo mejorará su inglés o escritura, y tambien ten
en cuenta que su participación no afectará sus notas en la clase regular. También, pueden
dejar de participar en este investigación cuando lo quieran. Cuando tenga usted cuestiones,
podría llamarme a (xxx)xxx-xxxx
! Yes, I am interested in learning more.
Sí, me gusta aprender más.
! No, I am not interested.
No, no lo quiero.
Date/Fecha:
____________________ __________________________
Student Parent/Guardian
Estudiante Padre/Guardián
137
Date of Preparation: [insert date]
USC UPIRB #
Expiration Date:
University of Southern California
Rossier School of Education
Waite Phillips Hall, 1001D
Los Angeles, CA 90089-4031
ASSENT FORM FOR RESEARCH
Page 1 of 2
ASSENT TO PARTICIPATE IN RESEARCH
An Action Research, Using Task-Based Instruction and Focus on Form
in Open Court Independent Work Time for Written Language
1. My name is Alexandros Bantis.
2. We are asking you to take part in a research study because we are trying to learn more about
different ways to teach writing for elementary English language learners
3. If you agree to be in this study, you will come to Room for about 30-40 minutes after school,
until about 3:45 p.m. (2:45 p.m. when school gets out early on Tuesdays). We will work on
writing for about a month (about 20 times). You will be writing paragraphs about the Open
Court stories. At least once per week you will meet with me one-on-one to do a writing
conference to make your writing easier to understand. We will make a final draft in the
writing conference, and then you will need to make corrections to the first draft by checking
the final draft we will make together. Most of the time you will be working independently. I
would like to make an audio recording of what we talk about so that I can understand how to
make the writing conference better, but if you don’t want me to tape it that’s okay.
4. If you don’t feel comfortable writing or talking about your writing, then you might feel
uncomfortable being a part of this study because we are going to do a lot of writing and
talking about writing. Also, when we do the writing conference, there will be other kids in
the room who are working independently, so maybe you could feel a little embarrassed if
they hear us talking about your writing. But you don’t have to talk if you don’t want to
5. I don’t know if your writing or English will get better. It isn’t going to change the grade your
teacher will give you, and won’t change whether you go to the next grade. You will get a
138
Page 2 of 2
Date of Preparation: [insert date]
USC UPIRB #:
Expiration Date:
granola bar at the beginning of each meeting so you can have more energy. And I will give
you a plastic fountain pen and bottle of ink (worth about $35) when you are done as a way of
saying, “Thank you” for your hard work.
6. Please talk this over with your parents before you decide whether or not to participate. We
will also ask your parents to give their permission for you to take part in this study. But even
if your parents say “yes” you can still decide not to do this.
7. If you don’t want to be in this study, you don’t have to participate. Remember, being in this
study is up to you and no one will be upset if you don’t want to participate or even if you
change your mind later and want to stop.
8. You can ask any questions that you have about the study. If you have a question later that
you didn’t think of now, you can call me (323)394-0399 or ask me next time.
9. Signing your name at the bottom means that you agree to be in this study. You and your
parents will be given a copy of this form after you have signed it.
It is OKAY to make tape recordings of the writing conferences.
It is NOT OKAY to make tape recordings of the writing conferences.
___________________________________ ____________________
Name of Subject Date
___________________________________
Subject’s Signature
___________________________________ ____________________
Name of Investigator Date
___________________________________
Name of Investigator
139
APPENDIX E: DEVELOPMENT OF CRITERIA FOR DATA CODING
Regarding the prompt linguistic focus, arriving at a set of criteria and
categories was problematic and difficult. The exact categorization evolved over the
course of the analysis. The initial categories of prompt focus consisted of
punctuation, verb tense, other grammar, coherency, and cohesion. Coherency was
defined as the degree of understandability while cohesion was defined as the degree
to which the sentences were consistent with the given topic sentences.
Very quickly, this categorization scheme for the prompt focus proved to be
inadequate. During this second stage of coding of the prompt focus, there came to be
the following categories: coherence, cohesion, contraction, end-mark, expansion,
meaning, miscellaneous, prepositions, pronouns, quote marks, complete sentence,
syntax, tense, and verb. Contraction was defined as prompts that ultimately ended
with some sentence or portion of sentences being deleted, while expansion was
defined as adding additional information to sentences or adding additional sentences.
This system proved to be cumbersome and did not yield meaningful information.
The final evolution of the coding system entailed looking at the data set from
the initial perspective, which was looking at barriers of the interlanguage to a fluid
and effortless transmission of meaning from the author to the reader. Celce-Murcia &
Olshtain (2000) discussed revising a writing piece from the perspective of Grice’s
(1975) maxims of quantity, quality, relevance, and manner. Issues of manner were
classified according the major grammatical categories of: punctuation (primarily
quotation marks and end marks), verb tense (one of the original targeted features),
140
other verb issues (primarily issues of using past continuous versus simple past, for
example), prepositions (both the use of prepositions and idiomatic usage within
phrasal verbs), syntax (this was in some cases a matter of word order or in some
cases it was that a critical element of the sentence was missing), and pronouns (in
some cases there was confusion due to an unclear antecedent and in other cases
confusion due to mixing objective/nominative cases of pronouns). A small number of
the turns were also classed as similes and semantics.
The framework of Grice’s maxims meant understanding the underlying
purpose of recasting the participant’s sentences with either more or less information.
The underlying issue was lack of a shared schema. Often times the participant
attempted to share some event but assumed that the reader had sufficient background
information to know what was going on, as the excerpt below indicates (ns=native
speaker, nns=non-native speaker):
Prompt: I remember a time I thought up a way to fix a problem.
Draft: n/a
Recast: Then I said, “You can only choose one element to play.
NS: okay, my question is ... I don’t understand, what was the problem?
NNS: that he choose three things.
NS: he choose three things ... okay so the problem was ... what was the
problem?
NNS: he choose three things.
NS: so the problem was that he chose three things, but he should have chosen ...
NNS: one
NS: AH!, okay, so can we say that then? ... because I didn’t really understand
the problem.
NNS: then I said, you can only choose two ... ONE element.”
NS: one element, you can only choose one element ...
NNS: to play.
NS: to play ... and that fixed the problem?
--Conference 09 / Turns 23-28
141
In some cases there was extraneous information added to the paragraph that
did not contribute to understanding the meaning, and the sentence (or portion of a
sentence) needed to be taken out in order to create sentences that remained focused
on providing support to the topic sentence. These were classified as Topicality
Prompt: I remember a time I used my imagination to play
NS: okay, so one question ... I was fighting evil people. I was throwing water
balls at them. ... so that is the part where you used your imagination to play,
right? ... and then this part right here Then it got so late that I had to go
inside my house. Is that talking about using your imagination to play?
NNS: no
NS: Then I smelled food cooking. It was egg with beans and soup. does that
have to do with your imagination and playing?
NNS: no
NS: okay, Next it was time to go to sleep. Does that have to do with your
imagination and play?
--Conference 29 / Turns 12-14
In some cases there was a fundamental lack of understanding of the prompt
and, thus, rather shallow topic development. These cases proved difficult to code.
Should these cases be categorized as semantics (lack of understanding of the
meaning of the topic sentence), schema (in this case a mismatch between the schema
of the teacher who created the topic sentence and the student who was writing the
detail sentences), or should it be coherency (that the student-created detail sentences
did not sufficiently develop the teacher-created topic sentence)? In the example
below, the participant did not understand the concept of imagination and wrote a
paragraph that was essentially devoid of meaning. Hence, it assumed the qualities of
a grammar exercise rather than a genuine communicative task.
142
Prompt: I will show my friends how to yose There imagination.
1draft: And I will show them.
2draft: I will show my friends how to yose There imagination.
3draft: I will show my wife how to use her imaginetion to.
4draft: I will hsow my hole family to.
--Conference 01 / Turns 5-6
Once the categorization scheme was in place, the last piece needed was to
quantify the amount of time spent within the conferences focusing on the different
targets. To do this, I decided to assign a value according to the length of the text
string generated from the turn. For example, the following comes from the 11th turn
of the first writing conference. It was a successful recast, and the purpose was verb
tense (using the simple future tense). The length of the text of this turn was 27
characters for the by me and 30 characters by the participant.
NS: I ... will, say I will go
NNS: I will go treasure hunting ...
By assuming that the length of the text strings within the transcription
reflected the actual amount of time spent talking during the conference, it was
possible to quantify the proportion of time spent on the various topics.
143

Abstract (if available)

Abstract The purpose of this study was to investigate the impact of task based writing instruction (TBwI) on English language acquisition and differentiated instruction for minority language students during the Independent Work Time instructional component of the Open Court Reading program. One teacher and 10 third grade students (8-9 years old) participated in this mixed methods study. TBwI was as a platform for communicative language teaching. Together they recast the students written interlanguage embedded into standard English. The study took place after school, 45 minutes per day for one month, resulting in 35 transcribed writing conferences, writing samples, and interviews. Results indicate that TBwI can be a useful vehicle for differentiated instruction, constructivist pedagogy, and principles of second language acquisition to address the diverse needs of second language learners.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Maximizing English learners' success in higher education with differentiated instruction

PDF

Effective reading instruction for English learners

PDF

Explicit instruction’s impact on the student achievement gap in K-12 English language learners

PDF

The role of music in the English language development of Latino prekindergarten English learners

PDF

The use of culturally relevant authentic materials and L1 in supporting second language literacy

PDF

Technology as a tool: uses in differentiated curriculum and instruction for gifted learners

PDF

Preparing English language learners to be college and career ready for the 21st century: the leadership role of secondary school principals in the support of English language learners

PDF

Factors associated with second language anxiety in adolescents from different cultural backgrounds

PDF

Building capacity in homeroom teachers to support English language learners

PDF

Instructional proficiency strategies for middle school English language learners

PDF

Status of teaching elementary science for English learners in science, mathematics, and technology centered magnet schools

PDF

Building academic vocabulary for English language learners through professional development: a gap analysis

PDF

Teach them to read: implementing high leverage pedagogical practices to increase English language learner academic achievement

PDF

Foreign-language teachers' needs to achieve better results: the role of differentiated instruction

PDF

English language learners utilizing the accelerated reader program

PDF

The use of differentiation in English medium instruction in Middle Eastern primary schools: a gap analysis

PDF

Narrowing the English learner achievement gap through teacher professional learning and cultural proficiency: an evaluation study

PDF

Critical features for teaching the five-paragraph essay to middle school Chinese speaking English learners

PDF

Teachers’ perceptions and implementation of instructional strategies for the gifted from differentiation professional development

PDF

Building teacher competency to work with middle school long-term English language learners: an improvement model

Asset Metadata

Creator Bantis, Alexandros Merkouris (author)

Core Title Using task based writing instruction to provide differentiated instruction for English language learners

School Rossier School of Education

Degree Master of Science

Degree Program Teaching English to Speakers of Other Languages

Publication Date 02/28/2008

Defense Date 01/16/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag constructivist pedagogy,differentiated instruction,OAI-PMH Harvest,Open Court Reading,Second language instruction,second language learning,task based instruction

Language English

Advisor Mora-Flores, Eugenia (committee chair), Filback, Robert (committee member), Shakhbagova, Julietta (committee member)

Creator Email abantis@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1026

Unique identifier UC1377862

Identifier etd-Bantis-20080228 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-41221 (legacy record id),usctheses-m1026 (legacy record id)

Legacy Identifier etd-Bantis-20080228.pdf

Dmrecord 41221

Document Type Thesis

Rights Bantis, Alexandros Merkouris

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu