Explicitly Teaching Strategies, Skills, and Knowledge: Writing Instruction
in Middle School Classrooms
Susan De La Paz
Santa Clara University Steve Graham
University of Maryland
Writing is a very demanding task, requiring the orchestration of a variety of cognitive resources. For
developing writers, it can be especially demanding, as they have not yet mastered important writing
processes, skills, and knowledge involved in planning, drafting, and revising text. In the present study,
middle school students were directly taught strategies that facilitated the execution of each of these
processes. They were also taught the knowledge and skills needed to carry out these strategies. In
comparison to peers in the control condition, students in the experimental treatment condition produced
essays that were longer, contained more mature vocabulary, and were qualitatively better. These gains
were evident immediately following instruction and on a short-term maintenance probe administered 1
Writing is one of the most difficult skills that children are
expected to master in school (Graham & Harris, in press). The
difficulty of learning how to write competently is reflected in data
collected as part of the National Assessment of Educational
Progress (Greenwald, Persky, Campbell, & Mazzeo, 1999). At the
4th-grade level, for example, slightly more than 60% of the stu-
dents tested were classified as “basic” writers, demonstrating only
partial mastery of the skills and knowledge needed at that grade
level. Another 16% of 4th-grade students scored below this basic
achievement level. Similar results were obtained at the 8th- and
12th-grade levels. These findings indicated that the average Amer-
ican student is not a proficient writer, as the percentages of
students who performed at or above the proficient level remained
approximately 25% at each grade level.
Concerns about children’s writing have led to calls for improve-
ments in the teaching of writing (e.g., Riley, 1996), with many
states and school districts designing and implementing procedures
to boost students’ writing performance (e.g., Bridge, Compton-
Hall, & Cantrell, 1997). This is a particularly challenging task,
however, as skilled writing requires the acquisition and coordina-
tion of strategies for regulating the writing process (e.g., strategies
for planning, monitoring, evaluating, and revising), skills for pro-
ducing text (e.g., handwriting, spelling, sentence construction),
and knowledge about specific genres, writing conventions, and so
forth (Graham & Harris, 2000; Hayes, 1996; Hayes & Flower,
1980; Scardamalia & Bereiter, 1986). Thus, efforts aimed at im-
proving writing instruction need to help students develop the
strategies, skills, and knowledge needed to write effectively.
The present study examined the effectiveness of an instructional
program designed to improve the writing performance of middle
school students. Although the program primarily focused on teach-
ing students strategies for planning, drafting, and revising text, the
knowledge and skills needed to support these processes were also
emphasized. This emphasis included knowledge about the charac-
teristics of good writing, criteria for evaluating writing, and the
structure of expository essays that involved explanation and per-
suasion (the writing task emphasized in this study). Writing skills
that were addressed included constructing a thesis statement and
using mature vocabulary, transition words, and different types of
sentences. These skills are not only important in constructing a
good essay but were also stressed on a state-wide writing compe-
tency exam taken by students.
The key element of the instructional program was a strategy that
organized and directed the processes for planning and writing an
essay. The students developed a plan in advance of writing that
involved analyzing the demands of the writing assignment, setting
goals for writing, and generating and organizing material to write
about. The strategy also prompted students to use their plan while
writing; revise and upgrade their plan as needed; and include
transition words, interesting or mature vocabulary, and varied
(error-free) sentence types.
We placed considerable emphasis on planning, because plan-
ning is a critical element in skilled writing. High levels of planning
are especially apparent in the composing behavior of skilled writ-
ers. For instance, Kellogg (1987) reported that college students
devote about one fourth of their writing time to planning, whereas
Gould (1980) indicated that business executives spend about two
thirds of their composition time planning. Skilled writers plan not
only what they will write but how they will write it, establishing
goals for their writing, structuring their ideas, and considering the
needs of their audience (Bereiter & Scardamalia, 1987; Flower &
In contrast, school-age children often minimize the role of
planning in writing, doing little deliberate planning, particularly
planning in advance (McCutchen, 1995; Scardamalia & Bereiter,
1986). It is not uncommon for students to start writing almost
immediately after a writing task is assigned, spending less than a
Susan De La Paz, Department of Education, Santa Clara University;
Steve Graham, Department of Special Education, University of Maryland.
Correspondence concerning this article should be addressed to Susan De
La Paz, Department of Education, 226 Bannan Hall, Santa Clara Univer-
sity, 500 El Camino Real, Santa Clara, California 95053. E-mail:
Journal of Educational Psychology Copyright 2002 by the American Psychological Association, Inc.
2002, Vol. 94, No. 4, 687–698 0022-0663/02/$5.00 DOI: 10.1037//0022-06126.96.36.1997
1 LINE LONG
minute planning their composition in advance (Burtis, Bereiter,
Scardamalia, & Tetroe, 1983). Even when prompted to plan in
advance, students’plans are often meager and relatively unsophis-
ticated (e.g., Bereiter & Scardamalia, 1987; Berninger, Whitaker,
Feng, Swanson, & Abbott, 1996; Boscolo, 1990). For example, the
prompted plans developed by seventh- through ninth-grade stu-
dents in a study by Berninger et al. (1996) typically involved a
listing of ideas or words to be used when writing.
The approach to planning emphasized in this study involved the
development of an advanced plan that can be modified and up-
graded as it is implemented. Advanced planning may be especially
advantageous for developing writers. A written plan provides an
external memory, where the youngster can store ideas without the
risk of losing them. It may further reduce the need to plan while
writing, freeing resources to engage in other writing processes,
such as translating ideas into words, transcribing words into
printed text, or reviewing and revising text (Kellogg, 1986, 1987).
Advanced planning can also provide a mechanism for addressing
some aspects of writing not commonly considered by youngsters
when planning while writing. Many students approach writing by
retrieving any information from memory that is topic appropriate
and writing it down, with each preceding phrase or sentence
stimulating the generation of the next idea (McCutchen, 1988;
Scardamalia & Bereiter, 1986). With this approach, little attention
is directed to the needs of the reader, constraints imposed by the
topic, or the organization of the text. One or more of these factors
can be emphasized during advanced planning.
In addition to the development of a flexible plan to guide the
process of drafting or writing a composition, considerable empha-
sis was placed on the process of revising. This emphasis included
encouraging students to revise their plans as needed and to make
revisions during writing that were designed to meet goals involv-
ing the use of interesting words, transitions words, and varied
sentences. As students learned to use the planning and writing
strategy, they also met with their peers to receive feedback on how
they could revise their essay to make it better. Students gave their
peer-editing partner feedback in the following areas: ideation,
organization, vocabulary, sentence structure, and grammar. In ad-
dition, teachers provided feedback on various aspects of the com-
position during individual conferences.
We decided to emphasize revising, which is similar to planning,
because it is also a critical element of skilled writing (Fitzgerald,
1987; Graham & Harris, 2000). For example, in a study of adult
writers who were asked to think aloud while composing, Hayes
and Flower (1980) found that 10% to 15% of the content state-
ments produced during composing involved the process of revis-
ing. Revising is especially apparent in the composing behavior of
professional writers. Edmund Morris, the biographer, indicated
that he rewrote each sentence six times, as he struggled to turn it
“into something lucid”(Lamb, 1997, p. 18). Raymond Carver, the
popular mystery writer, indicated that he did as many as 20 to 30
drafts of some stories (Plimpton, 1989). Like planning, revising
generally plays a much more limited role in the compositions
produced by developing writers. School-age children do not revise
frequently, extensively, or skillfully (Fitzgerald, 1987; Graham,
The self-regulated strategy development model (SRSD; Graham
& Harris, 1993; Harris & Graham, 1996) was used to teach the
writing strategies, skills, and knowledge targeted for instruction in
the present study. With SRSD, students are explicitly taught writ-
ing strategies along with procedures for regulating these strategies
and the writing process. These procedures include goal setting,
self-monitoring, and self-instructions. In addition, knowledge and
skills students need to use the target strategies and write effectively
are also taught.
In most studies of SRSD to date, the participants were students
with special needs, primarily children with learning disabilities. In
over 20 studies, planning and revising strategies have been suc-
cessfully taught to these children, resulting in improved writing
performance (see Harris & Graham, 1999). Only two investiga-
tions have examined the application of SRSD with normally de-
veloping writers. In a single-subject design study by Danoff,
Harris, and Graham (1993), the schematic structure and story
quality of 3 average writers (2 fifth graders and 1 fourth grader)
improved following instruction in the use of an advanced planning
strategy taught using SRSD. Somewhat similar results were ob-
tained in a second single-subject design study (De La Paz, 1999),
where the quality of 22 seventh- and eighth-grade students’writing
improved following SRSD instruction with the planning and writ-
ing strategy used in the present study. The present study extends
previous SRSD research by examining its effectiveness with nor-
mally achieving writers, using a more traditional large-group de-
sign. Furthermore, the participating students’regular teachers de-
livered instruction, providing evidence on the viability of this
approach in a traditional classroom setting. Although teachers
delivered instruction in several other SRSD studies (Danoff et al.,
1993; De La Paz, 1999, 2001; MacArthur, Graham, Schwartz, &
Shafer, 1995; MacArthur, Schwartz, & Graham, 1991; MacArthur,
Schwartz, Graham, Molloy, & Harris, 1996), most investigations
have involved the use of specially trained tutors providing either
one-on-one or small-group instruction.
Theoretical conceptualizations of skilled writing typically em-
phasize the complexity and the self-regulatory nature of compos-
ing (Flower & Hayes, 1980; Hayes, 1996; Zimmerman & Risem-
berg, 1997). Skilled writers activate and coordinate an impressive
array of mental operations, skills, and knowledge as they make
plans, draw ideas from memory, develop concepts, organize ideas,
create a written draft, reconceptualize plans, revise text, and so
forth (Graham & Harris, in press). Although the mechanisms
responsible for the development of such competence are not yet
fully understood, theories involving the development of expertise
suggest that the path from acclimation to competence is shaped by
changes in strategic and subject-matter knowledge (e.g., Alex-
ander, Graham, & Harris, 1998; Alexander & Judy, 1998; Bjork-
lund, 1990). An important assumption in the present study, there-
fore, is that teachers can help students become more competent
writers by directly teaching them the processes, skills, and knowl-
edge that underlie effective writing as well as how to coordinate
and regulate their use. Because such instruction addresses several
aspect of composing (i.e., planning, producing, and revising text as
well as regulating the writing process) that are challenging for
developing writers (Scardamalia & Bereiter, 1986), we expected
that the instructional program evaluated in this study would have
a positive impact on the participating students’writing perfor-
mance. This improvement included the development of more so-
phisticated writing plans as well as the creation of essays that were
longer, contained more mature vocabulary, and were qualitatively
better than the essays produced by students in the control
688 DE LA PAZ AND GRAHAM
Setting and Participants
Setting. The study took place in two middle schools in a suburban
school district in the Southeast. The two middle schools had similar
demographic characteristics. One school had a population of 504 students;
94% of the students in the school were White, 5% were African American,
and 1% were Asian or Hispanic. Approximately 18% of the students in the
school received free or reduced-fee lunch prices. Less than 1% of the
students received services for English as a second language. The second
school had a population of 540 students, with identical percentages of
students in each of the previous categories. The only exception was that
12% of the students in the second school received free or reduced-fee lunch
Teachers. Five seventh- and eighth-grade teachers agreed to partici-
pate in this study. We randomly assigned the 10 language arts classes
taught by these teachers to the experimental and control conditions as
follows: 6 were randomly assigned to the experimental condition (i.e.,
strategy instruction). Three of these classes were at the eighth-grade level
and 3 of these classes were at the seventh-grade level. We assigned the 4
remaining classes to the control condition (with 2 classes at each grade
level). Thus, teachers taught both experimental and control classes.
It is important to note that one of the control classrooms was subse-
quently dropped from the study, as the writing program used in this
classroom differed significantly from the other control classes. Although
this teacher provided time for writing (as did the other teachers), he did not
provide instruction in how to compose an essay. This instruction was done
by the other control teachers. It is unlikely that the removal of his class-
room from the study had an appreciable impact for two reasons. One,
only 2 of the 30 original control students were in this teacher’s classroom.
Two, these 2 students were not markedly different from the other control
students in terms of their characteristics or entering writing skills.
The four remaining teachers all held teaching credentials in education,
and three of them had completed a bachelor’s degree in education as well.
The 2 eighth-grade teachers had taught for 15 and 26 years, respectively,
whereas the 2 seventh-grade teachers had taught for 2 and 15 years,
respectively. All of the teachers were White women, and each was blind to
the theoretical issues and hypotheses of the study.
Students. The students in this investigation included 58 seventh and
eighth graders; 30 students participated in the experimental condition
and 28 students participated in the control condition. Whereas classes had
been randomly assigned to the experimental and/or control conditions,
students were drawn from intact classes, resulting in a quasi-experimental
design. Each student was in either the strategy instruction condition or in
the control condition; thus, no student participated in both forms of
instruction. The 58 students described in this study were part of a larger
population; however, we decided to analyze a representative subset of data
because a detailed analysis of the written products from the entire popu-
lation was prohibitive. The steps used to select and assign students to the
two conditions are described following.
First we eliminated any student who had been in an instructional re-
search study the previous year. Second, students who were absent 3 or
more days during instruction were dropped from the larger pool. This
judgment was determined by examining attendance records from the
school offices. Third, we applied a stratified random sampling procedure,
using proportional allocation to ensure the total number of students from
each teacher in our final sample was proportional to the total number of
students taught by each teacher in the original population. Our rationale
was that teachers each had one or more language arts classes each day. As
a result, we preserved the ratio of students to teachers in our final sample,
as in the larger population. The exact procedures were followed within
each condition. The percentage of participating students from each class
ranged from 8% to 26% (differences in percentages were due in part to the
initial elimination of students who had participated in the previous research
study and were absent 3 or more days). Finally, none of the participating
students received special education services.
The students’most recent percentile ranks on the Reading and Language
Arts portions of the Comprehensive Tests of Basic Skills (CTBS; 1989), a
group-administered achievement battery, were available for the majority of
students (N⫽25 in both the experimental and control conditions). This test
is given every spring, typically during March and April, as one of the
high-stakes measures mandated by the state. On this test, percentile ranks
between 25–75 are considered average. The majority of students in each
condition had average or greater-than-average reading skills according to
this test (97% in the experimental condition and 89% in the control
condition). Language arts skills were also average or above for nearly all
students (93% in the experimental condition and 100% in the control
Information on the characteristics of students by condition is presented
in Table 1. There were no significant differences among students assigned
to the two conditions in terms of grade, Pearson
(1, N⫽58) ⫽2.56, p⫽
.11, Crame´r’sV⫽.21; chronological age, F(1, 56) ⫽0.88, MSE ⫽57.93,
p⫽.35 (effect size ⫽0.24); CTBS reading percentile ranks, F(1,
48) ⫽1.80, MSE ⫽968.00, p⫽.19 (effect size ⫽0.37); or CTBS
Language Arts percentile ranks, F(1, 48) ⫽1.17, MSE ⫽776.18, p⫽.29
(effect size ⫽0.32).
In addition, at pretest, students’plans and essays were compared to
determine whether the two groups differed significantly. Four separate
one-way analysis of variance (ANOVA) tests were conducted to evaluate
the relationship between the two instructional conditions and planning (i.e.,
the level of development of students’prewriting plans), essay length (i.e.,
number of words written), vocabulary (i.e., number of words seven letters
or longer), and overall holistic quality (see Tables 2 and 3 for means and
standard deviations for these measures). At pretest, the planning behavior
of students in the two groups was not significantly different, F(1,
56) ⫽0.87, MSE ⫽4.34, p⫽.77 (effect size ⫽–0.06). Students assigned
to the two conditions also wrote papers that were essentially the same
length, F(1, 56) ⫽0.00, MSE ⫽12.81, p⫽.95 (effect size ⫽0.02), with
the same number of novel words seven letters or longer, F(1, 56) ⫽0.58,
MSE ⫽23.68, p⫽.45 (effect size ⫽0.21), and similar in overall quality,
F(1, 56) ⫽0.12, MSE ⫽9.49, p⫽.74 (effect size ⫽–0.09).
Materials. The composition task chosen for investigation in this study
was expository essays that involved explanation and persuasion. This
particular genre was selected over other forms of expository writing (e.g.,
Summary of Student Characteristics by Condition
SD 7.94 8.26
SD 22.72 23.67
SD 27.20 24.31
Male 18 20
Female 12 8
7th 19 23
8th 11 5
Note. CTBS ⫽Comprehensive Test of Basic Skills.
WRITING INSTRUCTION IN MIDDLE SCHOOL
compare and contrast) because the participating teachers wanted to prepare
students for a state-wide writing competency exam, emphasizing this type
of writing. A pool of five topics previously used by Susan De La Paz
(1999) was shown to two graduate students, both previous teachers, who
selected three prompts judged to be similar in terms of interest and
difficulty. The three topics were as follows: (a) “Think about how students
can improve their grades, and write an essay telling how students can
improve their grades and give reasons explaining why you think these
actions are important;”(b) “Think about rules that you think are not fair,
and write an essay telling what rules should be changed and give reasons
explaining why you think these changes are important;”and (c) “Think
about things a true friend does, and write an essay telling what a true friend
does and give reasons explaining why you think these things are impor-
tant.”To control for confounding due to differences in student interest,
Summary of Procedures for Experimental and Control Groups
Preinstruction Students were taught the characteristics of
expository essays. Students were taught the characteristics of expository
Students were taught the basic (5-paragraph) essay
form. Students were taught the basic (5-paragraph) essay form.
Students were taught terminology for understanding
writing prompts. Students were taught terminology for understanding writing
Pretesting Single 35-min session to plan and compose an
essay. Single 35-min session to plan and compose an essay.
Similarities Students composed 5 essays (the same essay topics
were used in both conditions, and 35 min were
allotted for composing each essay).
Students composed 5 essays (the same essay topics were
used in both conditions, and 35 min were allotted for
composing each essay).
Students selected 1 of their essays for a class
portfolio. Students selected 1 of their essays for a class portfolio.
Teachers reviewed the different types of sentences. Teachers reviewed the different types of sentences.
Students were retaught the basic (5-paragraph)
essay form. Students were retaught the basic (5-paragraph) essay form.
Differences Students were taught to independently use the
PLAN and WRITE strategy to compose an essay
(this included establishing the purpose and the
benefits of the strategy and its steps, describing
it, modeling its use, memorizing the mnemonic
and the strategy steps, and providing adjusted
teacher and peer assistance until the strategy
could be applied independently).
Teachers directed the generation and organization of
writing ideas, using either webbing or Power Writing,
before students composed essays (students were
periodically reminded to use these procedures when
writing the essay).
Students were taught the knowledge and skills
needed to use the PLAN and WRITE strategy
(this included instruction in composing a thesis
sentence and introductory paragraph; use of
varied sentence types, interesting vocabulary, and
transition words; maintaining control of the
topic; and procedures for evaluating the quality
of an essay).
Students were taught a variety of discrete writing skills
(this included vocabulary, spelling, and grammar
Students were taught to use self-regulatory
procedures to facilitate the acquisition and the
use of the PLAN and WRITE strategy (this
included goal setting, self-monitoring, self-
evaluation, and self-instructional statements).
Teachers graded students’essays, providing them with
feedback on the quality of their writing (essays were not
revised by students, however).
Students were provided with temporary procedural
support to help them initially use the PLAN and
WRITE strategy (this included a brainstorming
sheet for organizing ideas as well as an essay
sheet and cue cards that reminded them to
include basic essay and paragraph components).
Students read assigned books during class time.
Students revised their essays after receiving
feedback from a peer (this was facilitated by
teaching them how to use a revising checklist to
provide and receive feedback).
Students were provided homework to reinforce skills taught
Students were administered a quiz to test their
knowledge of the steps of the PLAN and WRITE
Students were administered tests and quizzes to assess their
knowledge of material taught in class.
Posttesting Single 35-min session to plan and compose an
essay. Single 35-min session to plan and compose an essay.
Maintenance One month after posttesting, a single 35-min
session to plan and compose an essay. One month after posttesting, a single 35-min session to
plan and compose an essay.
690 DE LA PAZ AND GRAHAM
familiarity, or knowledge, the administration of the three essay topics were
counterbalanced across students and probes (pretest, posttest, and mainte-
nance) using a Latin square design.
Table 2 provides a visual outline of the instruction for students in the
experimental and comparison group. The preinstruction session as well as
the pretest, posttest, and maintenance testing sessions are included as well.
Before administration of the writing pretest, all students (control and
experimental) participated in a single preinstruction session lasting an
entire class period. We did this to ensure that each student was familiar
with basic attributes of expository writing. To expose students to the
writing task, teachers used a scripted lesson on the characteristics of good
essays and a basic essay form (i.e., five paragraphs with a thesis statement,
supporting paragraphs, and conclusion). Students were told that expository
essays inform, explain, clarify, define, or instruct by giving information,
explaining why or how, clarifying a process, or defining a concept. Teach-
ers explained that the format of the writing prompt comprised two ele-
ments. The first part was to help students think about the subject that they
were to write about (e.g., “Think of a person that you would like to meet.
This person may be from the present or the past”), and the second part gave
exact writing directions (e.g., “Write an essay telling whom you would like
to meet and give specific reasons for wanting to meet him or her”).
Students were told that they were expected to write a five-paragraph essay;
the decision to use this structure had been reached by consensus in
conversations with teachers before the study began. Teachers then de-
scribed the basic purpose of each paragraph (i.e., to introduce, to provide
a main idea and supporting details, etc.), and explained unfamiliar terms
(e.g., thesis statement). After the preinstruction session, students in both
conditions completed the pretest in a single 35-min session. Each student
received a sheet of paper that included the assigned prompt as well as space
to record notes.
Following the pretest, students completed the 6-week instructional phase
of the study. During this instructional period, students in both conditions
wrote an equal number of essays (in response to the same prompts) and
were allotted the same amount of time (35 min) to write each essay.
Students in the experimental condition used the planning, writing, and
revising procedures they were taught to compose two essays collabora-
tively (in large and small groups) and to compose three essays indepen-
dently. Students in the comparison condition planned and wrote one essay
as a whole-class activity and wrote four additional independent essays.
Once instruction was completed, and before the posttest, students in both
conditions selected a finished essay for his or her class portfolio.
After the instructional phase of the study, students in each condition
completed a posttest using the same procedures as the pretest. For 1 month
following the end of instruction, teachers did not assign any composition
tasks. At the end of this period, each student was administered a mainte-
nance writing probe, again using identical procedures to determine whether
the effects of treatment were maintained over time. After completing the
short-term maintenance probe, Susan De La Paz interviewed teachers to
obtain information about perceived effectiveness of the instructional pro-
cedures, recommendations, and other comments.
The participating teachers received an instructor’s manual with scripted
lesson plans and other instructional materials to guide their teaching of the
writing program. They learned how to teach the program using the SRSD
model of instruction in two full-day workshops. A rationale for each SRSD
instructional component was presented along with an overview of the
strategies, skills, and knowledge taught. This overview included (a) pur-
pose and description of the planning and writing strategy; (b) activating
background knowledge of basic essay parts, use of synonyms, and use of
different types of sentences; (c) reviewing students’initial writing abilities;
(d) modeling the planning strategy; (e) implementing collaborative (whole-
class and small-group) practice; (f) providing guidance to students when
using the peer-revision checklists; (g) fading teacher support during inde-
pendent practice; and (h) mastery criteria.
The first day of the workshop was primarily used (a) to provide an
overview of the planning and writing strategy, (b) to model the strategy,
and (c) to explain goal setting and how to fade instructional support as
students gained mastery. The second day focused on the teachers practicing
how to deliver the program and how to address problems that might arise.
Specific Procedures for Experimental and Comparison
Experimental group. Students in the experimental condition learned
specific strategies for planning, drafting, and revising text. They were also
taught skills and knowledge needed to apply these strategies as well as to
write effectively. The strategies, skills, and knowledge were taught using
the SRSD model (Harris & Graham, 1996). SRSD shares important fea-
tures with other types of instruction in which strategies and supporting
skills and knowledge are taught (cf. Deshler & Schumaker, 1986; Englert
et al., 1991; Schumaker & Deshler, 1992; Wong, 1997; Wong, Butler,
Ficzere, & Kuperis, 1997) in that teachers provide think-aloud demonstra-
tions followed by verbal scaffolding as a vehicle for helping students gain
independence in using the target strategies.
In contrast, and central to the SRSD model, students also learn proce-
dures for regulating use of the strategy and the writing task. Self-regulatory
procedures typically include goal setting, self-instructions, and self-
monitoring. SRSD incorporates support and components aimed at helping
students deal with affective challenges such as poor motivation, impulsiv-
ity, devaluation of learning, as well as cognitive difficulties with memory
and other aspects of information processing, low task engagement and
persistence, and low productivity (De La Paz, Owen, Harris, & Graham,
2000; Harris & Graham, 1996). Thus, a critical feature of SRSD is that
teachers typically model and help students identify verbal statements and
physical actions to promote student mastery of the targeted writing process.
Six instructional stages provide the framework for SRSD. Procedures for
promoting maintenance and generalization are integrated throughout in-
struction. The six stages of instruction in the SRSD model are develop
background knowledge,describe it,model it,memorize it,support it, and
independent performance (Harris & Graham, 1996). Each of these stages is
italicized and placed in parentheses when it occurs in the following
discussion of the writing program. Instructional procedures for students
paralleled the outline used to prepare the participating teachers; however,
what was presented to teachers in 2 days was spread out over a 6-week
Percentage of Students Engaging in Specific Types of Planning
Type of planning
1 234 5
Pretest 80 17 0 3 0
Posttest 3 3 3 77 13
Maintenance 0 7 3 50 40
Pretest 80 10 7 3 0
Posttest 23 40 7 27 3
Maintenance 8 23 3 37 28
Note. Percentages do not always add up to 100 because of rounding.
Planning scores are as follows: 1 ⫽no advanced planning; 2 ⫽listing
ideas; 3 ⫽topics given with emerging subordination; 4 ⫽three subtopics,
each with two or more details; 5 ⫽accurate, fully developed map or
WRITING INSTRUCTION IN MIDDLE SCHOOL
grading period for students. Teachers who participated in a related study
during the previous year had requested an increase in time from 4 weeks to
the present 6 weeks of instruction to allow students more opportunity to
revise essays and to take short breaks in writing successive essays. Con-
sequently, teachers provided the strategy instruction 4 days per week,
allowing for 1 day per week for other content.
Each stage of the SRSD model involved one or more instructional
sessions. The first lesson provided an overview about the purpose as well
as a description of the first part of the planning and writing strategy
(describe it). This overview included a discussion on how writers use
planning strategies when they write, the benefits of using such strategies,
and the goal for learning this strategy (to write a good essay). Teachers then
introduced the planning and writing strategy and the rationale for each step.
The mnemonics PLAN and WRITE were used to help students remember
strategy steps and also served as a reminder to plan before starting to write
and to continue the planning process while writing. Students took notes on
strategy steps and rationale throughout the discussions.
It is important to note that specific elements in the strategy were
designed to match the task (writing on demand) and perceived goals (based
on the state’s scoring rubric) for the high-stakes writing assessment ad-
ministered by the state in which the participating schools were located.
Thus, the first part of the strategy, PLAN, focused on interpreting the
writing prompt and developing a writing plan in advance of writing. The
first step, “Pay attention to the prompt,”helped students fully consider the
topic by identifying (a) what they were being asked to write about and (b)
how they should develop their essay. This step was seen as of critical
importance, because previous writing assignments on the state’s profi-
ciency exam varied considerably over past years. To illustrate, consider the
prompt, “Choose the age you would like to be. Write an essay explaining
why you want to be this age.”Students were taught to underline once what
they were being asked to write about (tell an age one wished to be) and to
underline twice how they were asked to develop it (give reasons explaining
why this age was chosen). In comparison, a different prompt required
students to “Think about a special event you will never forget. Write an
essay telling what happened, how you felt, and why it is unforgettable.”To
answer the second prompt, students had to identify a memorable event and
to develop his or her essay by explaining what happened, how the writer
felt, and what made the event memorable.
During the second step, “List main ideas,”the writer first decided on one
topic (such as which age or which event to write about) and then brain-
stormed at least three main ideas for the development of the essay (three
reasons why this age was of interest). To encourage reflection in this step,
instructors asked students to brainstorm three responses to the prompt
before deciding on one to write about. In the third step, “Add supporting
ideas,”students added at least three details, examples, or elaborations that
supported each main idea. As part of this step, students were encouraged to
revise their chosen topic in instances when they could not brainstorm an
adequate number of supporting ideas. The fourth step, “Number your
ideas,”was included to remind students to arrange their main ideas in the
order they planned to use them. This step also encouraged students to be
purposeful in the development of their topics. Whereas for some topics the
order of one’s main ideas was arbitrary, for other topics this sequence could
signal literary tension or importance of topic.
The second part of the strategy, WRITE, was designed to remind
students to use their plan, continue the planning process while writing, and
address writing goals or plans set during a prewriting conference. Step 5,
“Work from your plan to develop your thesis statement,”reminded students
to incorporate ideas from their plan into a thesis statement. With this step,
students were shown how to compose both basic and advanced introduc-
tory paragraphs, by varying the placement of the thesis statement. To meet
the needs of more capable writers, students were shown that an alternative
to presenting the thesis statement as the first sentence was to “start with an
attention getter,”or lead up to their thesis statement in one of the following
ways: (a) use a series of questions or statements, (b) use a brief or funny
story, (c) use a mean or angry statement, or (d) start with the opposite
opinion from what you believe. In general, eighth-grade students were
shown both approaches, and seventh-grade students were shown the more
advanced approach on an individual basis after demonstrating mastery with
the basic approach to writing introductory paragraphs.
Step 6, “Remember your goals,”was an additional reminder to meet
writing goals set during an individual conference, prior to writing. These
goals were written to reflect qualities of good writing, and to meet specific
criteria reflected by the rubric designed by the state for scoring writing.
Goals included maintaining control of the topic, organization, use of
mature vocabulary, and use of varied (error-free) sentence types. As
students neared mastery of the target strategy, teachers helped students set
individualized goals, focusing on one or two areas of need for each student,
based on the quality of compositions written during independent practice.
Steps 7 through 9 of the strategy gave specific suggestions for planning
while writing. These three steps were to “Include transition words for each
paragraph”;to“Try to use different kinds of sentences”; and to use
“Exciting, interesting, $100,000 words”in their compositions. These sug-
gestions were intended to serve as additional prompts to help students
reflect on the state’s criteria for proficient essays. Students also learned that
once they finished their essays, they should keep these criteria or steps in
mind as they reviewed their essay. Students were encouraged to check that
they used transition words appropriately, to make minor changes (e.g.,
creating a compound sentence from adjacent simple sentences), to add
variety to their paper, and to substitute synonyms for words that were used
more than once in a given paragraph.
After describing the PLAN and WRITE strategy, teachers reviewed
students’knowledge of basic essay parts and operationalized the goals
included in the last four steps of WRITE. Teachers used additional sessions
to provide students with information about their initial planning and
writing abilities and to introduce the idea that writing plans need to be
revised and upgraded during writing (develop background knowledge). To
accomplish the first objectives, teachers presented two model essays on
successive days. They showed students selected writing prompts, reviewed
the first step of PLAN, and then engaged the class in a discussion about the
model essays. On the first day, students read the essay and labeled the
introductory, body, and concluding paragraphs. Then they examined
whether the model essay contained a good thesis statement, identified
transition words in the paragraphs, and discussed ways to improve the
model essay. Students also searched the essay for different sentence types
according to form (simple, compound, and complex) and function (declar-
ative, imperative, exclamatory, and question). The second day added
vocabulary use as an additional element of focus. Students identified what
they considered to be “exciting, interesting, $100,000 words”in the model
essays and made suggestions for additional changes in vocabulary. Teach-
ers also conducted whole-class reviews of the strategy steps.
During the next few sessions, teachers explained the holistic scoring
rubric used by the state, emphasizing how it matched what they were
learning. Teachers then engaged in brief individual conferences with each
student to describe how they would evaluate the youngster’s essays,
identifying one or two areas in need of improvement. Students were
encouraged to select one or two goals that would help remedy target areas
of weakness. Because the classes included approximately 30 students,
teachers gave students an assignment to work on while they waited to have
their conferences. The assigned task was to practice the first step in the
PLAN and WRITE strategy (underline once what you are being asked to
write about and underline twice how you will develop the essay) on a
worksheet with 12 prompts. Students who finished early then brainstormed
ideas for one or more writing prompts and the whole class again reviewed
the strategy. Portions of several subsequent sessions were needed to discuss
how students responded to this worksheet.
The next stage of instruction required each teacher to demonstrate how
to use the PLAN and WRITE strategy by thinking aloud when planning
and writing an essay (model it). It required up to 3 days to complete this
interactive demonstration. While modeling the planning strategy, the in-
structor used a variety of self-instructions to show how she managed the
692 DE LA PAZ AND GRAHAM
strategy and the writing process. These included self-statements involving
problem definition (e.g., “Since I decided to put my thesis statement first,
I will write it as the beginning of my introductory paragraph”), planning
(e.g., “OK, my next step is to...”), self-evaluation (e.g., “I’m off to a
good start”), and coping (e.g., “I really like what I’ve come up with so
While the instructor modeled the strategy, she also demonstrated how to
use three different types of procedural support for using the PLAN and
WRITE strategy. The first support was a brainstorming sheet for organiz-
ing ideas. This form prompted the writer to first identify possible responses
to the prompt, and it included blank lines in outline form for writing main
ideas and details. The second support was referred to as an “essay sheet,”
and it prompted students to first write their thesis statement and then to
decide whether to place the thesis statement as the first or last sentence in
the introductory paragraph. This form also provided indented lines and
subheadings (e.g., “body paragraph two”or “conclusion”) to remind stu-
dents about essay parts and basic paragraph writing skills. The third
support was a series of cue cards (De La Paz, 1999) designed to be a final
reminder of what each paragraph should include.
Throughout modeling, the recursive use of various processes and pro-
cedures was emphasized. For example, during Steps 2 and 3 of PLAN,
“List main ideas ”and “Add supporting details,”the teacher generated
additional main ideas as well as details, examples, and elaborations and
selected three main ideas (and elaborations) from the pool of ideas.
Similarly, during the last two steps of WRITE, “Try to use different kinds
of sentences”and “Exciting, interesting, $100,000 words,”the teacher
made revisions as she was in the process of writing the paper.
Modeling was followed by several days of guided instruction, where
teachers helped students use the PLAN and WRITE strategy to plan and
compose a class essay. Students then worked in small groups to plan and
compose a second essay (support it). Teachers prompted students to
complete each step of the strategy. Students learned to write “PLAN”on
the top of their brainstorming sheets and “WRITE”on the top of their essay
sheets to guide their use of the strategy (and to cross off each letter as steps
were completed) and to note one or two goals on the top of the page that
focused on problematic areas. Whole-class discussions focused on (a) the
relationships between introductory and concluding paragraphs, (b) review
of different sentence types, (c) identification of different sentence forms,
and (d) examples of mature vocabulary in students’essays.
In addition, as is regularly the case in process-writing classrooms,
students were provided time to revise their papers. This was accomplished
in two ways. First, students routinely engaged in peer-revising conferences.
To introduce this task, teachers reviewed partner rules for sharing and
revising essays and modeled partner-revising procedures to ensure that
students worked cooperatively. A checklist developed for this purpose is
shown in the Appendix. Teachers found the need to monitor students’use
of the checklist, however, to prevent students from automatically giving
each other only positive responses (i.e., when improvement was
Second, teachers met with students throughout the periods devoted to
small-group and independent work to discuss difficulties students had in
applying the PLAN and WRITE strategy and providing comments regard-
ing student progress and goal setting. Whereas students were initially
encouraged to use the same goals for planning and composing, as soon as
students began to work independently (see following paragraph), teachers
became more diagnostic in their suggestions for each student. This in turn
had helped teachers manage the way they provided feedback to students
regarding their writing progress. Teachers evaluated whether students met
their goals and chose additional goals that could be selected for subsequent
During the final stage of instruction (independent performance), students
used the PLAN and WRITE strategy to write essays but received needed
assistance from the instructor in applying it. Throughout this stage, teach-
ers purposefully reduced their assistance, thus shifting responsibility for
using the strategy to the student. Prompting and guidance, collaborative
goal setting, use of brainstorming, essay sheets, cue cards, and the time
available for planning and writing were gradually reduced. At this point,
students learned to assume the additional responsibility of independently
setting goals and to develop essay plans and subsequent compositions on
regular sheets of notebook paper. Students were encouraged to think
independently about new writing prompts, and teachers systematically
reduced the number of times they assisted students on each subsequent
attempt at using the planning and writing strategy. Each student engaged in
approximately 5 days of independent practice (writing up to four essays).
Criterion performance required students to (a) use the strategy indepen-
dently (without relying on the brainstorming or essay sheet or cue cards),
(b) generate one plan and essay within the 35-min time period, and (c)
master at least one self-determined goal. After completing the independent
practice period, students selected one essay to be part of a class portfolio.
As indicated previously, students verbally rehearsed the steps for PLAN
and WRITE (memorize it) throughout the 6-week instruction. Teachers
gave brief, whole-class review on the mnemonics and their steps, and when
students completed work early, they were paired for small-group review. In
addition, during or after independent practice, each teacher gave students a
formal assessment of their learning. Students recalled the strategy sequence
and underlying rationale for each step, provided examples of transition
words and synonyms, and described how to write an introductory
Control condition. Students in the control condition received a rela-
tively traditional writing curriculum (see Table 2). This curriculum in-
cluded providing time in class to write compositions, teaching the mechan-
ics of writing (e.g., spelling, grammar, and vocabulary), teacher direction
and guidance in generating and organizing ideas for writing essays, as-
signing homework, and periodically testing students about the skills and
knowledge taught in the classroom. Grammar, spelling, vocabulary, and
writing a composition were typically taught as distinct skills, but students
were expected to have improved grammar, vocabulary, and spelling skills
in their compositions.
The control condition was further structured by asking teachers to use
the same essay topics that were used in the experimental treatment condi-
tion (see Table 2). Thus, control students were provided with the same
prompts for writing essays and were given the same number of opportu-
nities and amount of time to compose them. Like their experimental
counterparts, students in the control condition also selected a single essay
to include in a class portfolio, reviewed the basic types of sentences, and
were retaught the structure for the five-paragraph essay form (this had
initially been taught in the preinstruction lesson). The structure for the
five-paragraph essay form was retaught using either the materials from the
preinstructional lesson or by using comparable materials from the teachers’
language arts texts.
Although the three control classrooms were similar in many respects,
they were not identical. For example, they did not teach exactly the same
mechanical skills from one class to another (e.g., one teacher did not
explicitly teach spelling, whereas the others did). Likewise, one of the
teachers did not assign homework, but the others did. Teachers in control
classrooms also helped students generate and organize ideas for writing
essays, conducting this as a whole-class activity. In two classrooms,
teachers directed the generating and organizing of writing ideas using a
semantic web, whereas ideas were generated and organized using Power
Writing (Sparks, 1988) in the other control classroom. Although it might
have been preferable to have control classrooms that were identical, this is
extremely difficult to achieve in a school setting, especially when a study
involves more than one grade.
It is important to note that none of the control students were taught the
strategies included in the experimental treatment (e.g., PLAN and WRITE
or the peer-revising checklist). Likewise, the SRSD model was not used in
the control classrooms. Although teachers in the control classes helped
students write essays by directing the generation and organization of
writing ideas (using either webbing or Power Writing) as a whole-class
activity, there was little attempt on the part of teachers to help students
WRITING INSTRUCTION IN MIDDLE SCHOOL
internalize these cognitive processes beyond direct demonstration of their
application and periodic reminders to use them. This relatively weak form
of direct instruction (see Sawyer, Graham, & Harris, 1992) differs consid-
erably from the cognitive strategy instruction provided to students in the
experimental group, where the purpose and rationale of the target strategies
were established, their application was modeled, students received exten-
sive and adjusted assistance (including the use of procedural supports such
as think sheets and cue cards) in learning to apply them independently, and
a variety of self-regulation procedures (e.g., goal setting, self-monitoring,
etc.) were used to support their acquisition and use. Although it might have
been preferable to ask teachers not to emphasize planning at all during the
course of instruction, this would have created an ethical problem given the
length of the study and the importance of planning in writing (Flower &
Hayes, 1980; Gould, 1980; Kellogg, 1987).
Finally, there was limited overlap in the type of skills or knowledge
taught to students in the two conditions. Children in both conditions were
retaught the five-paragraph essay format and reviewed basic sentence
types. Students in the control condition were also taught a variety of basic
writing skills (e.g., spelling, vocabulary, and grammar) as separate ele-
ments of the writing process. In contrast, the teaching of skills in the
experimental condition was integrated around learning to use the PLAN
and WRITE strategy, as students were taught the skills and knowledge
(e.g., transition words) they needed to successfully use this strategy.
In summary, the control condition primarily involved instruction in the
basic format for writing a five-paragraph essay, practice in writing such
essays, teacher help in generating and organizing essay ideas, discrete
instruction on a variety of basic writing skills, and homework and tests
designed to support the learning of these skills. Although students in the
experimental condition also received instruction in the five-paragraph-
essay writing format and practiced writing such essays, they learned to
independently use an integrated set of strategic processes, skills, and
knowledge for developing such essays.
We implemented the following safeguards to ensure that the experimen-
tal instructional procedures occurred as planned. First, teachers who taught
the experimental treatment program received an instructor’s manual with
lesson plans and practiced teaching until they could implement them
successfully. Second, at the end of each week, a graduate student familiar
with the purpose and design of the study randomly selected 20% of the
audiotapes for fidelity checks. The graduate student checked off steps from
the instructor’s manual when listening to strategy instruction lessons. On
all but one occasion, the graduate student checked off 100% of the steps as
completed. The only exception involved a single period in which one of the
teachers implemented only 56% of the strategy steps. Third, a second
graduate student made weekly visits to observe instruction in both condi-
tions, making notes and reporting problems to Susan De La Paz when
needed. For example, one teacher failed to provide cue cards to students
unless they initiated a request for them (i.e., limiting her support for their
use). Another teacher did not remember to fade out the use of cue cards
(i.e., limiting the potency of the final independent practice session). Any
teacher not implementing the experimental treatment program with 100%
accuracy was given written corrective feedback to follow. Fourth, Susan
De La Paz made weekly visits to all teachers during their planning time to
note scheduling changes and discuss any concerns. Typical issues related
to not having enough time to instruct students in identifying what the
prompt was about, needing new supplies, and so forth.
All teachers recorded daily activities throughout the instructional period
on a data collection sheet designed for this purpose. The same graduate
student who made weekly visits to observe instruction in the experimental
classrooms did so as well for the control classes, and she generated written
summaries of the day’s events when watching control lessons. In addition,
the graduate student who listened to audiotapes from the experimental
treatment lessons also randomly selected 20% of the control audiotapes for
fidelity checks. This individual created written summaries of the control
lessons. Both types of summaries and the teachers’daily notes were used
to document the instruction in the control condition. The summaries and
daily written notes were analyzed inductively to create categories that
captured the instructional activities or moves of the control teachers.
Table 2 provides a summary of the instructional activities that occurred in
the control classrooms during the 6-week instructional period.
Planning. On the basis of procedures by Whitaker, Berninger,
Johnston, and Swanson (1994), all written plans received a score for level
of development. Plans were scored on a 5-point scale, ranging from 1 (no
advanced planning)to5(accurate, fully developed). Both degree of
elaboration (by mapping or outlines) and accuracy (demonstrated by an-
swering the prompt and logical subordination of main ideas and details)
were used to evaluate planning. Students’plans were given the highest
rating if they generated a map or outline identifying a central theme in
response to the prompts with three main ideas and two or more logically
related examples for each main idea. In addition, to earn this rating, the
student had to respond to all parts of the writing prompt accurately, neither
omitting requisite information nor adding irrelevant content. Conversely, a
list of propositions without subordination or a plan that was the same as a
first-time draft received a score of 2. Susan De La Paz scored all plans, and
a former middle school teacher unfamiliar with the purpose and design of
the study independently scored a random sample of 20% of the plans.
Interrater reliability (agreements divided by agreements plus disagree-
ments) was .81.
Length. All essays were scored in terms of number of words written.
We counted all words that represented a spoken word regardless of spell-
ing. A graduate student unfamiliar with the purpose or design of the study
counted the number of words in all essays. A second graduate student
unfamiliar with the purpose and design of the study independently counted
a random sample of 20% of the essays (r⫽.96).
Vocabulary. We scored all essays for the number of words that
included seven or more letters. This procedure provided an index of the
maturity of students’vocabulary. This rating has been used in stan-
dardized assessments of writing (cf. the Test of Written Language;
Hammill & Larsen, 1988). In scoring this measure, we counted only the
first instance of a given word that was greater or equal to seven letters
to reduce the chance of inflating a student’s score on the basis of
redundant rather than varied vocabulary use. In addition, we corrected
misspelled words, incorrectly hyphenated words, and eliminated non-
words such as inschools before scoring. Susan De La Paz scored all
essays, and a former teacher who was unfamiliar with the purpose and
design of the study independently counted a random sample of 20% of
the essays. Interrater reliability (agreements divided by agreements plus
disagreements) was .87.
Quality. A traditional holistic rating scale was used to assess quality.
Two former middle school teachers who were unfamiliar with the design
and purpose of the study independently scored all essays. Essays were rated
from a low score of 1 to a high score of 8, representing the reader’s general
impression of overall quality. We told each teacher to read each paper
attentively, but not laboriously, and to take into account quality of ideas,
structure of text, word choice, sentence structure, and mechanics—not
assigning undue weight to any one factor. Anchor points were established
by selecting a high essay (score of 7), a middle essay (score of 4), and a low
essay (score of 1). These essays were obtained from seventh- and eighth-
grade classrooms that were in the target schools but not included as part of
the study. The readers practiced scoring additional papers that were not part
of the data set until they could use the anchor points to reliably score
papers. The Pearson product⫺moment correlation for quality was .87.
Only one rating differed by more than a single point. Differences were
resolved by averaging scores.
694 DE LA PAZ AND GRAHAM
Table 3 presents planning scores for students in each group at
pretest, posttest, and maintenance. The means and standard devi-
ations for length, vocabulary, and quality for the two instructional
conditions at pretest, posttest, and maintenance are presented in
A 2 (instructional group) ⫻2 (trials) repeated measures
ANOVA design was used to evaluate the relationship between the
instructional conditions (experimental treatment and control) and
students’planning to determine whether planning scores differed
significantly at posttest and maintenance. Table 2 presents descrip-
tive information. The trials effect was statistically significant, F(1,
56) ⫽8.60, MSE ⫽3.13, p⬍.01, whereas the interaction effect
was not. It is important to note that the Fratio for the main effect
for instructional conditions was significant, F(1, 56) ⫽30.61,
MSE ⫽60.10, p⫽.00 (posttest effect size ⫽1.17; maintenance
effect size ⫽1.04). Examination of Table 2 reveals that the
majority of students (i.e., 80%) did no advanced planning on the
pretest. On the posttest and maintenance measures, however, the
majority of students in both conditions created written plans, but
the plans of students in the experimental treatment condition were
better developed. Ninety percent of these students created plans
that received a score of 4 or 5 at posttest and maintenance. In
contrast, only 30% and 65% of control students’posttest and
maintenance plans, respectively, received equivalent scores.
A 2 (instructional group) ⫻2 (trials) repeated measures
ANOVA design was used to evaluate the relationship between the
instructional conditions (strategy instruction and control) and stu-
dents’total essay length to determine whether these scores differed
significantly at posttest and maintenance. Table 3 presents descrip-
tive information. The trials and the interaction effect were not
statistically significant; however, the Fratio for the main effect for
instructional conditions was significant, F(1, 56) ⫽11.04,
MSE ⫽65,812.87, p⬍.01 (posttest effect size ⫽0.82; mainte-
nance effect size ⫽1.07). Thus, after instruction ended and 1
month later, students in the experimental treatment group wrote
papers that were significantly longer than students in the control
A 2 (instructional group) ⫻2 (trials) repeated measures
ANOVA design was used to evaluate the relationship between the
instructional conditions and the total number of novel vocabulary
words that were seven letters or longer in students’essays to
determine whether these scores differed significantly at posttest
and maintenance (see Table 3). The trials effect was statistically
significant, F(1, 56) ⫽4.49, MSE ⫽80.69, p⫽.04, whereas the
interaction effect was not. It is important to note that the Fratio for
the main effect for instructional conditions was significant, F(1,
56) ⫽13.41, MSE ⫽925.43, p⬍.01 (posttest effect size ⫽1.13;
maintenance effect size ⫽0.94). Immediately following instruc-
tion, students in the experimental treatment condition wrote papers
with a greater number of different words that were seven letters or
longer when compared with students in the control condition.
Furthermore, these gains were maintained 1 month after instruc-
A 2 (instructional group) ⫻2 (trials) repeated measures
ANOVA design was used to evaluate the relationship between the
instructional conditions and the holistic rating of essay quality at
posttest and maintenance (see Table 3). The trials and the inter-
action effect were not statistically significant; however, Fratio for
the main effect for instructional groups was significant, F(1,
56) ⫽17.42, MSE ⫽13.53, p⫽.00 (posttest effect size ⫽1.71;
maintenance effect size ⫽0.74). After instruction and 1 month
later, students in the experimental group wrote papers that were
judged to be of higher overall quality than students in the control
This study examined the effects of an integrated writing pro-
gram on middle school students’essay-writing abilities. The in-
tervention applied in this study taught students strategies for plan-
ning, drafting, and revising an expository essay involving
explanation and persuasion. Students were also taught the knowl-
edge and skills needed to carry out these strategies as well as to
write effectively. Students learned to apply the inculcated pro-
cesses, skills, and knowledge in a flexible but coordinated manner.
We anticipated that such instruction would have a salutary impact
on students’writing performance, because it addressed aspects of
composing (i.e., planning, producing, revising text as well as
self-regulation in the writing process) that are challenging for
developing writers (Scardamalia & Bereiter, 1986).
As expected, the writing program had a positive effect on the
writing performance of the participating middle school students.
Immediately following instruction, students in the experimental
group produced essays that were longer, contained more mature
vocabulary, and were qualitatively better than the essays generated
Means and Standard Deviations for Length, Vocabulary, and
Quality by Condition
Pretest 184.33 59.02 183.39 63.92
Posttest 236.17 67.43 190.82 55.61
Maintenance 229.67 77.01 179.68 46.80
Pretest 13.60 6.73 12.32 6.03
Posttest 17.23 8.03 11.17 5.36
Maintenance 18.50 6.89 13.25 5.60
Pretest 3.13 0.94 3.21 0.88
Posttest 3.63 0.89 2.86 0.45
Maintenance 3.73 1.01 3.14 0.80
WRITING INSTRUCTION IN MIDDLE SCHOOL
by youngsters in the control classrooms. These effects were main-
tained on an essay written 1 month after instruction ended. It is
important to note that the impact of the experimental treatment was
strong, as effect sizes for the three writing product measures
(length, vocabulary, and quality) ranged from 0.82 to 1.71 on the
posttreatment and maintenance writing probes. These findings are
consistent with a small body of literature, showing that a multi-
component program centered around writing-process instruction
can improve the writing performance of developing writers
(Danoff et al., 1993; De La Paz, 1999; Englert et al., 1991; Yeh,
A key element of the writing program was a planning strategy
(i.e., PLAN) that helped students analyze the demands of the
writing assignment and generate and organize possible writing
content. Prior to the start of instruction, 80% of the students in both
the experimental and control conditions did not generate any
written plans in advance of writing. Following instruction, written
plans were much more common for both groups of students, with
97% of the experimental students generating an advanced plan and
77% of the control students doing the same. Nevertheless, the
written plans generated by students in the experimental condition
were better developed than those produced by youngsters in the
control condition. The plans of the students in the experimental
condition tended to be more complete, elaborate, and hierarchical
than those produced by their counterparts in the control condition.
These differences were sustained on the short-term maintenance
probe administered 1 month after instruction ended. As with the
writing-product measures, the impact of the experimental treat-
ment on students’written plans was quite strong, as effect sizes
were greater than 1.00 on both the posttreatment and maintenance
Additional research is needed to replicate the present findings
and to determine what aspects of the treatment were responsible
for changes in students’writing behavior. It is likely that the
intensive and highly explicit strategy instruction that formed the
core of this writing program is responsible for a sizable proportion
of these gains. Previous research has shown that strategy instruc-
tion in either planning or revising improves students’writing
performance (e.g., Beal, Garrod, & Bonitatibus, 1990; De La Paz
& Graham, 1997; Harris & Graham, 1999; Wong, Butler, et al.,
1997; Zimmerman & Risemberg, 1997). It is further likely that the
writing skills and knowledge students were taught contributed to
their gains in writing. The development of competence is shaped
not only by enhanced strategic behavior but by changes in knowl-
edge and skills as well (Alexander et al., 1998).
Application of SRSD in the Regular Classroom
The present study provides additional verification that the SRSD
model can be applied successfully with developing writers in
traditional classrooms. Most of the students who have participated
in studies using the SRSD model were students with writing
disabilities, and they often received instruction from a specially
trained tutor (see Harris & Graham, 1999). The present study,
along with two previous investigations (Danoff et al., 1993; De La
Paz, 1999), demonstrate that SRSD also provides an effective
means for teaching normally developing writers and that this
instruction can be delivered in the regular classroom. Teachers in
this and the two prior investigations were able to administer SRSD
with a high degree of fidelity, and students receiving such instruc-
tion made substantial gains in writing.
Because of the increasing emphasis on inclusion (Gartner &
Lipsky, 1987), it has become even more important to identify
instructional practices that are successful with a wide range of
children. Inclusion is likely to be more acceptable and less bur-
densome to teachers if they have access to instructional procedures
that benefit normally developing students as well as those with
disabilities. SRSD provides one instructional approach that is
effective with both groups. For example, the PLAN and WRITE
strategy taught in this study improved the writing performance of
normally developing writers as well as students with learning
disabilities, attention-deficit disorders, and speech and language
difficulties in two previous studies (De La Paz, 1999, 2001).
The present study provides support for the educational practice
of directly teaching writing strategies, along with the skills and
knowledge needed to apply them. Other studies (Danoff et al.,
1993; De La Paz, 1999; Englert et al., 1991; Yeh, 1998) have also
shown that such instruction improves the writing performance of
normally developing students. In these studies, teachers modeled
how to use writing strategies (and accompanying skills and knowl-
edge), providing extensive instruction, practice, and assistance
until students could apply them independently. This method differs
considerably from the classroom practices of many teachers in
which there is little evidence of extended or even explicit instruc-
tion in writing strategies and many writing skills (Anthony &
Anderson, 1987; Graham & Harris, 1996).
Why is explicit and more extended instruction in writing not
more prominent in schools? One reason is because many teachers
rely on informal or incidental methods of learning (i.e., the natural
learning approach) to promote writing development (see Graham
& Harris, 1997). Proponents of the natural learning approach
believe that explicit instruction is not necessary, as strategies,
skills, and knowledge are best learned through real use in mean-
ingful and authentic contexts. Several studies, however, challenge
this assumption, by showing that students who participate in writ-
ing programs based on these principles benefit when more exten-
sive, structured, and explicit instruction is included (Danoff et al.,
1993; MacArthur et al., 1991; Yeh, 1998). For example, Danoff et
al. (1993) found that children in process-writing classrooms, where
informal and incidental learning methods were emphasized, made
considerable gains in writing when SRSD instruction in planning
was provided through an extended series of minilessons.
Another possible reason why explicit instruction in writing is
not more prominent is that some educators believe that directly
teaching writing strategies, skills, and knowledge is counterpro-
ductive and may even be harmful (Elbow, 1981; Freedman, 1993;
Petraglia, 1995). In the area of planning, for instance, preplanning
is viewed by some experts as potentially harmful because it at-
tempts to circumvent the nonlinear and recursive nature of writing
and keeps writers from exploiting ideas or opportunities that might
otherwise arise while writing (see Kellogg, 1990). Others argue
that explicit teaching is counterproductive because rhetorical be-
havior is not rule governed or independent of the context in which
it is applied (Petraglia, 1995). As this and other studies (Danoff et
al., 1993; De La Paz, 1999; Englert et al., 1991; Harris & Graham,
1999; Yeh, 1998) have demonstrated, however, such instruction is
696 DE LA PAZ AND GRAHAM
beneficial. Following explicit strategy instruction, for example,
students’knowledge of writing increases, they use more sophisti-
cated writing strategies, and the quality of their writing improves
(Graham & Harris, 1993).
Alexander, P., Graham, S., & Harris, K. R. (1998). A perspective on
strategy research: Progress and prospect. Educational Psychology Re-
view, 10, 129–154.
Alexander, P., & Judy, J. (1998). The interaction of domain-specific and
strategic knowledge in academic performance. Review of Educational
Research, 58, 375–404.
Anthony, H., & Anderson, L. (1987, April). The nature of writing instruc-
tion in regular and special education classrooms. Paper presented at the
annual meeting of the American Educational Research Association,
Beal, C., Garrod, A., & Bonitatibus, G. (1990). Fostering children’s revi-
sion skills through training in comprehension monitoring. Journal of
Educational Psychology, 82, 275–280.
Bereiter, C., & Scardamalia, M. (1987). The psychology of written com-
position. Hillsdale, NJ: Erlbaum.
Berninger, V., Whitaker, D., Feng, Y., Swanson, L., & Abbott, R. (1996).
Assessment of planning, translation, and revising in junior high students.
Journal of School Psychology, 34, 23–52.
Bjorklund, D. (Ed.). (1990). Children’s strategies: Contemporary views of
cognitive development. Hillsdale, NJ: Erlbaum.
Boscolo, P. (1990). The construction of expository text. First Lan-
guage, 10, 217–230.
Bridge, C., Compton-Hall, & Cantrell, S. (1997). Classroom writing prac-
tices revisited: The effects of statewide reform on writing instruction.
Elementary School Journal, 98, 151–170.
Burtis, P., Bereiter, C., Scardamalia, M., & Tetroe, J. (1983). The devel-
opment of planning in writing. In G. Wells & B. Kroll (Eds.), Explora-
tions in the development of writing (pp. 153–174). Chichester, England:
Comprehensive Tests of Basic Skills (4th ed.). (1989). Monterey, CA:
Danoff, B., Harris, K. R., & Graham, S. (1993). Incorporating strategy
instruction within the writing process in the regular classroom: Effects
on the writing of students with and without learning disabilities. Journal
of Reading Behavior, 25, 295–319.
De La Paz, S. (1999). Self-regulated strategy instruction in regular educa-
tion settings: Improving outcomes for students with and without learning
disabilities. Learning Disabilities Research & Practice, 14, 92–106.
De La Paz, S. (2001). Teaching writing to students with attention deficit
disorders and specific language impairment. Journal of Educational
Research, 95, 37–47.
De La Paz, S., & Graham, S. (1997). Strategy instruction in planning:
Effects on the writing performance and behavior of students with learn-
ing disabilities. Exceptional Children, 63, 167–181.
De La Paz, S., Owen, B., Harris, K., & Graham, S. (2000). Riding Elvis’
motorcycle: Using self-regulated strategy development to plan and write
for a state writing exam. Learning Disabilities Research and Prac-
tice, 15, 101–109.
Deshler, D. D., & Schumaker, J. B. (1986). Learning strategies: An
instructional alternative for low-achieving adolescents. Exceptional
Children, 52, 583–590.
Elbow, P. (1981). Writing with power. New York: Oxford University Press.
Englert, C., Raphael, T., Anderson, L., Anthony, H., Steven, D., & Fear, K.
(1991). Making writing and self-talk visible: Cognitive strategy instruc-
tion writing in regular and special education classrooms. American
Educational Research Journal, 28, 337–373.
Fitzgerald, J. (1987). Research on revision in writing. Review of Educa-
tional Research, 57, 481–506.
Flower, L., & Hayes, J. (1980). The dynamics of composing: Making plans
and juggling constraints. In L. Gregg & E. Steinberg (Eds.), Cognitive
processes in writing (pp. 31–50). Hillsdale, NJ: Erlbaum.
Freedman, A. (1993). Show and tell? The role of explicit teaching in the
learning of new genres. Research in the Teaching of English, 27,
Gartner, A., & Lipsky, D. (1987). Beyond special education: A quality
system for all students. Harvard Educational Review, 57, 367–395.
Gould, J. (1980). Experiments on composing letters: Some facts, some
myths, and some observations. In L. Gregg & E. Steinberg (Eds.),
Cognitive processes in writing (pp. 97–127). Hillsdale, NJ: Erlbaum.
Graham, S. (1997). Executive control in the revising of students with
learning and writing problems. Journal of Educational Psychology, 89,
Graham, S., & Harris, K. R. (1993). Self-regulated strategy development:
Helping students with learning problems develop as writers. Elementary
School Journal, 94, 169–181.
Graham, S., & Harris, K. R. (1996). Self-regulation and strategy instruction
for students who find writing and learning challenging. In M. Levy & S.
Ransdell (Eds.), The science of writing: Theories, methods, individual
differences, and applications (pp. 347–360). Mahwah, NJ: Erlbaum.
Graham, S., & Harris, K. R. (1997). It can be taught, but it does not develop
naturally: Myths and realities in writing instruction. School Psychology
Review, 26, 414–424.
Graham, S., & Harris, K. R. (2000). The role of self-regulation and
transcription skills in writing and writing development. Educational
Psychologist, 35, 3–12.
Graham, S., & Harris, K. R. (in press). Literacy: Writing. In A. Roheena
(Ed.), Encyclopedia of cognitive sciences. Oxford, England: Nature
Greenwald, E., Persky, H., Campbell, J., & Mazzeo, J. (1999). National
assessment of educational progress: 1998 writing report card for the
nation and the states. Washington, DC: U.S. Department of Education.
Hammill, D., & Larsen, S. (1988). Test of Written Language—2. Austin,
Harris, K. R., & Graham, S. (1996). Making the writing process work:
Strategies for composition and self-regulation. Cambridge, MA:
Harris, K. R., & Graham, S. (1999). Programmatic intervention research:
Illustrations from the evolution of self-regulated strategy development.
Learning Disability Quarterly, 22, 251–262.
Hayes, J. (1996). A new framework for understanding cognition and affect
in writing. In M. Levy & S. Ransdell (Eds.), The science of writing:
Theories, methods, individual differences, and applications (pp. 1–27).
Mahwah, NJ: Erlbaum.
Hayes, J., & Flower, L. (1980). Identifying the organization of writing
processes. In L. Gregg & E. Steinberg (Eds.), Cognitive processes in
writing (pp. 3–30). Hillsdale, NJ: Erlbaum.
Kellogg, R. (1986). Designing idea processors for document composition.
Behavior Research Methods, Instruments, & Computers, 18, 118–128.
Kellogg, R. (1987). Effects of topic knowledge on the allocation of
processing time and cognitive effort to writing processes. Memory &
Cognition, 15, 256–266.
Kellogg, R. (1990). Effectiveness of prewriting strategies as a function of
task demands. American Journal of Psychology, 103, 327⫺342.
Lamb, B. (1997). Booknotes: America’s finest authors on reading, writing,
and the power of ideas. New York: Times Books.
MacArthur, C., Graham, S., Schwartz, S., & Shafer, W. (1995). Evaluation
of a writing instruction model that integrated a process approach, strat-
egy instruction, and word processing. Learning Disability Quarterly, 18,
MacArthur, C., Schwartz, S., & Graham, S. (1991). The effects of a
reciprocal peer revision strategy in special education classrooms. Learn-
ing Disability Research and Practice, 6, 201–210.
MacArthur, C., Schwartz, S., Graham, S., Molloy, D., & Harris, K. R.
WRITING INSTRUCTION IN MIDDLE SCHOOL
(1996). Integration of strategy instruction into a whole language class-
room: A case study. Learning Disability Research and Practice, 11,
McCutchen, D. (1988). “Functional automaticity”in children’s writing: A
problem of metacognitive control. Written Communication, 5, 306–324.
McCutchen, D. (1995). Cognitive processes in children’s writing: Devel-
opmental and individual differences. Issues in Education: Contributions
from Educational Psychology, 1, 123–160.
Petraglia, J. (1995). Writing as an unnatural act. In J. Petraglia (Ed.),
Reconceiving writing: Rethinking writing instruction (pp. 79–100).
Mahwah, NJ: Erlbaum.
Plimpton, G. (1989). The writer’s chapbook. New York: Viking Press.
Riley, R. (1996). Improving the reading and writing skills of America’s
students. Learning Disability Quarterly, 19, 67–69.
Sawyer, R., Graham, S., & Harris, K. R. (1992). Direct teaching, strategy
instruction, and strategy instruction with explicit self-regulation: Effects
on the composition skills and self-efficacy of students with learning
disabilities. Journal of Educational Psychology, 94, 340–352.
Scardamalia, M., & Bereiter, C. (1986). Written composition. In M. Wit-
trock (Ed.), Handbook of research on teaching (3rd ed., pp. 778–803).
New York: MacMillan.
Schumaker, J. B., & Deshler, D. D. (1992). Validation of learning strategy
interventions for students with LD: Results of a programmatic research
effort. In B. Y. L. Wong (Ed.), Contemporary intervention research in
learning disabilities: An international perspective (pp. 22–46). New
Sparks, J. E. (1988, November). Read/write for personal and economic
growth. Paper presented at the annual meeting of the American Asso-
ciation for Adult and Continuing Education, Tulsa, OK.
Whitaker, D., Berninger, V., Johnston, J., & Swanson, H. L. (1994).
Intraindividual differences in levels of language in intermediate grade
writers: Implications for the translation process. Learning and Individual
Differences, 6, 107–130.
Wong, B. (1997). Research on genre-specific strategies for enhancing
writing in adolescents with learning disabilities. Learning Disability
Quarterly, 20, 140–159.
Wong, B. Y. L., Butler, D. L., Ficzere, S. A., & Kuperis, S. (1997).
Teaching adolescents with learning disabilities and low achievers to
plan, write, and revise compare-and-contrast essays. Learning Disabil-
ities Research and Practice, 12, 2–15.
Yeh, S. (1998). Empowering education: Teaching argumentative writing to
cultural minority middle-school students. Research in the Teaching of
English, 33, 49–84.
Zimmerman, B., & Risemberg, R. (1997). Becoming a self-regulated
writer: A social-cognitive perspective. Contemporary Educational Psy-
chology, 22, 73–101.
Directions: Read each item. Place a ⫹next to each line that describes the essay as it is written now, and a ? next to each
line where the writer may need to make changes.
Ideas and Development
Fully addresses the topic (answers all parts of the prompt).
Good development of ideas, with many details elaborated and extended.
Presents details–examples in a way that helps the reader understand the topic.
Ideas are clear and well illustrated.
Organization, Unity, and Coherence
Topic is clearly identified.
Remains on topic.
Well organized, with a smooth flow from one idea to the next.
Clear introduction, body paragraphs, AND conclusion.
Uses transitions skillfully to link sentences or paragraphs together.
Good word choices that are appropriate, specific, and varied.
Uses synonyms appropriately.
The essay is fun to read OR tells the reader something about the writer’s personality.
Sentence Structure, Grammar, and Usage
Includes many different kinds of sentences (various lengths and structures).
Has few (or no) errors in grammar or word usage, and the essay is easy to read.
Received November 15, 2000
Revision received January 11, 2002
Accepted January 11, 2002 䡲
698 DE LA PAZ AND GRAHAM