Perceptions Of Writing Skills

Hunter M. Breled

Robert J. Jones

College Entrance Examination Board, New York

ABSTRACT

A random sample of 806 essays was taken from over 80,000 essays
written for the College Board's English Composition Achievement Test (ECT)
during December 1979. Using a special taxonomy of 20 writing characteristics,
these essays were subjected to a second special reading in September 1980 to
determine which of these 20 writing characteristics most influenced judgments
of writing quality. Among other analyses, scores developed for the quality of each
of the 20 writing characteristics were used to "post diet" holistic scores on the
same essays obtained for the regular ECT administration. The results showed
that certain characteristics of discourse, in contrast to syntactic and lexical
characteristics, influenced judgments the most. The characteristics of discourse
included organization, transition, use of supporting evidence, and the originality
of ideas presented. In the sample examined, traditional syntactic emphases-such
as subject-verb agreement, punctuation, and pronoun usage-had less influence
on scores assigned. The results suggest that instruction in English composition
courses should emphasize discourse skills.

INTRODUCTION

This is a report of a research investigation into the specific characteristics

of brief, impromptu essay writing. It is based on a national sample of essays
evaluated by national samples of English teachers and professors. The sample of
essays was obtained from the December 1979 administration of the College
Board's Achievement Test in English Composition, which includes an essay
written during a 20-minute period of time. A total of 806 essays was randomly
sampled from over 85,000 essays written by students applying to colleges that
required College Board Achievement Tests. These colleges are, for the most part,
more selective than the average college. Therefore, the students represented in
the sample of essays examined are, taken as a whole, above average. The
English teachers and professors who read the essays were from two groups. The
first group consisted of the readers who were assembled for scoring the ECT
essay holistically as a regular part of the December ECT administration. Those
readers come from all parts of the country and represent both secondary and
post-secondary institutions. The second group of readers were 20 college
professors, sampled nationally, who participated in a special holistic and analytic
reading of 806 of the same essays. The purpose of the project was to begin to
explore the characteristics of the writing of entering college students as
perceived by teachers and professors of English, generally, and to attempt to
isolate and describe important elements of the type of essay studied. Other
research in progress will use the data base developed to examine new
approaches to the assessment of writing skill.

Background of the Problem

A thesis of this investigation is that there exist a number of major

obstacles that operate in our society that obstruct the development of writing
skills. Research, as well as casual observation, suggests that writing is a complex
skill mastered only through lengthy, arduous effort. Few students in high schools
or in universities, and even fewer adults, ever approach the level of effort and
time needed to learn to write well. In the average college, instruction in English
composi tion is usually restricted to the freshman year. Past this point in the
educational cycle, there are relatively few opportunities to learn skills that were
not imparted earlier. For those who manage to enter graduate or professional
schools, there exist some further opportunities, but only a handful of such
schools have strong writing programs. Another major obstacle to effective writing
instruction is that many English teachers are ill-equipped to teach writing and
feel uncomfortable in their efforts to do so. Many programs that prepare English
teachers concentrate heavily on literature and give limited attention to writing.
There is even less emphasis on the teaching of writing, a skill that, like writing
itself, can be learned only through practice. In minority and urban learning
settings, these ftrst twoobstacles-insufficient teacher time and training-are
exacerbated . A_ third major obstacle to the improvement of writing skill ts that
teachers often disagree in their perceptions of it. Even though English teachers
at times agree on writing standards and criteria (Diederich 197 4 ), they do not
agree on the degree to which any given criterion should affect the score they
assign to a paper. When scoring the same set of papers-even after careful
instruction in which criteria are clearly defined and agreed upon-teachers assign
a range of grades on any given paper (Godshalk, Swineford, and Coffman 1966).
The standards that define good writing apparently do not serve well when scores
are assigned to a piece of writing. Research to date suggests that the lack of
agreement in evaluating student writing has two aspects: teachers do not agree
on how much weight to assign particular writing characteristics in scoring any
one essay nor on how much weight each characteristic should be given generally
(Thompson 1976). The student, moreover, needs to know more than a score in
order to focus his or her efforts in learning to write better.

The Need for Descriptions of Writing

When it is considered that the average student is exposed to only a limited

set of teachers, it is clear that what the student learns about writing may be very
limited and biased.
Direct and Indirect Assessment

The English subject specialists at Educational Testing Service customarily

distinguish between direct and indirect measurement of writing skills. The direct
measurement of 2 writing skills or ability employs the written composition or
essay.

The Development of Essay Questions for Testing Purposes

At Educational Testing Service, essay questions for testing purposes are

developed over a long period of time-at least a year and a half and often two
years.

Previous Research of a Similar Nature

We have already referred to some previous research on English

composition that is in some ways similar to that undertaken for the present
project.

PROCEDURES

To answer questions that past research has not clearly resolved and to
achieve other objectives stated, the ECT administration given in December of
each year was viewed as an important primary source of information.

Sampling

The fust task in the sampling design was to identify those examinees for
whom complete data on critical variables was available.

Preparation of Sampled Essays

Following sampling of the essays, they were prepared for a second reading
by removing students' names, sexes, and non-essential identification numbers,
printing two highquality copies, and assembling reading packets containing 40
essays (nominally) each.

-Figure 1. The PWS Taxonomy

The Taxonomy

A listing of writing characteristics was fust formulated by the Humanities staff in

the Test Development section of the College Board division of Educational Testing
Service.

Figure 1. A description of each of the 20 characteristics follows.

1. Statement of thesis and purpose (explicit or implicit).

2. Overall organization.
3. Rhetorical strategy.
4. Noteworthy ideas: originality of thought or insight.
 5. Use of supporting materials: examples, etc.
6. Writer's voice: tone and attitude.
7. Paragraphing and transition.
8. Variety of sentence patterns.
9. Sentence logic. "Sentence logic"
10.Pronoun usage: agreement with antecedent, case, etc.
11.Subject-verb agreement, verb forms, etc.
12.Pararell structure: clauses, series, etc.
13.Idiomatic usage: prepositions, phrases, etc.
14.Punctuation.
15.Use of modifiers.
16.Appropriate levels of diction.
17.Range of vocabulary.
18.Precision of diction, conciseness of phrasing, redundancy.
19.Use of figurative language
20.Spelling

The Essay Evaluation Form

the taxonomy was established, the next task was to design an instrument
for collecting data on teachers' judgments of the essays with respect to each
element of the taxonomy.

The Special Reading

The special reading of the 806 sampled essays was held on September 27,
1980, at Educational Testing Service offices in Princeton.

Variable Development

Following the special reading, the essay evaluation forms were coded, and
other ratings were made by staff at ETS.

Analyses

Analytical details are described in subsequent sections, but a brief

overview of these methods is useful at this point.
 DATA DESCRIPTION

Table 1 presents a comparison of the samples with the total population in

terms of mean scores on key variables.

Discourse Characteristics

Means, ranges, and standard deviations for the discourse characteristics

of the essays sampled for this study are presented as the first nine variables of
Table 2. The means and ranges are indicative of the tendencies of readers to
perceive particular characteristics either positively or negatively.

Table 1. Population and Sample Means for Selected Variablei'

Table 2. Discourse, Syntactic, and Lexical Characteristic Variable Data
Descriptiona
Table 3. Mechanical, Composite, Holistic and Test Score Data Descriptiona

Syntactic Characteristics

Table 2 also includes a basic description of the syntactic characteristics

data (variables l 0 through 15).

Lexical Characteristics

Variables 16 through 20 of Table 2, the lexical characteristics, follow a

pattern similar to that of the syntactic characteristics except that there is slightly
more variation, as indicated by the standard deviations and the ranges.

Mechanical Scores

A basic description of the mechanical scores is given at the top of Table 3

as variables 21 through 25.

Composite Scores

With the exception of PWS mechanical, composite scores were developed

as simple additions of variable scores.

Holistic Scores

Variables 31, 32, and 33 of Table 3 provide a basic description of the

holistic scores developed.

Test Scores

The means for the test scores, based on the four groups sampled suggest
that, overall, the samples represent slightly more academic skill than has been
represented in other samples.
 CORRELATIONAL ANALYSES

Appendix E contains detailed matrices of intercorrelations for all variables. This

section will consider only selected relationships.

Table 4. Correlations between Holistic Essay Scores and Characteristics of

Discourstf
Table5. Correlations between Holistic Essay Scores, and Syntactic, Lexical
and Mechanical Characterimcgll
Table 6. Correlations among Holistic Essay Scores, Composite Scores, and
Test Scoresa

Summmary of Correlational Analyses

Correlations between holistic essay judgments and all other variables were
examined for pooled and weighted samples.

MULTIPLE REGRESSION ANALYSES

The relationships just discussed in terms of simple correlations may be

better understood when variables are analyzed simultaneously in multiple
regression analyses because, although relationships may be observed, whether
these are independent of other relationships is not clear from simple correlations.
Multiple correlations allow for a determination of independent contributions to
prediction.

Table 7. Multiple Prediction of ECf Holistic Score from PWS Scores0

Predicting Holistic Scores from PWS Scores

We first examine the 20 characteristics of the PWS taxonomy and the five
mechanical scores as predictors of the holistic judgments made of the ECT
essays for the ECT administration in December 1979.

Table 8. Multiple Prediction of ECf Holistic Score from ECf Objective Score
and PWS Scoresa

Predicting Holistic Scores from ECT Objective and PWS Scores

The preceding analyses show that a limited set of the discourse

characteristics, even when judged in a relatively crude manner, provide an
estimate of holistic scores assigned at an earlier time by a different group of
readers.

Predicting Holistic Scores from All Significant Contributors

Having analyzed the contributions of the various PWS characteristics, the

ECT objective score, and the mechanical scores within categories of predictor
types, it is next of interest to examine contributors when predictors are
combined across categories.
 Table 9. Multiple Prediction of ECT Holistic Scorea from Significant
Contributors
Table 10. Correlations between ECT Holistic Essay Score and PWS
Characteristics for Four Groups

Summary of Multiple Regression Analyses

In the exploratory analyses of this section, the focus has been on the
isolation of a limited set of writing characteristics that influence most
significantly the judgments of brief, impromptu essays written by students
applying to some of the best American colleges.

GROUP COMPARISONS

As previously noted in the section describing procedures, essays for four groups
were randomly sampled from the population of over 85,000 students who took
the ECT examination in December 1979.

Correlational Comparisons

Complete matrices of intercorrelation for all four groups and all variables
given are given in Appendix E. The correlational comparisons presented here
focus on the ECT holistic score and relationships with the other variables
available for each group sampled.

Multiple Regression Comparisons

In a manner similar to that used for the pooled samples previously, we

may determine which of the PWS scores contribute significantly to the prediction
of the ECT holistic score when entered in multiple sets.

Other Sample Comparisons

It is of value to look beyond correlational relationships and regression

analyses to gain a different perspective of the data.

Table 11. Correlations between ECT Holistic Essay Score and PWS
Mechanical Scores, PWS Composites, Holistic Scores, and Test Scores for Four
Groups
Table 12. Multiple Prediction of ECT Holistic Score from PWS Characteristics
for Black Sample (N=202)
Table 13. Multiple Prediction of ECf Holistic Essay Score from PWS
Characteristics for Hispanic (N) Sample (N=200)
Table 14. Multiple Prediction ofECT Holistic Score from PWS Characteristics
for Hispanic (Y) Sample (N=202)
Total Population Comparisons

For a limited set of variables available for all persons who took the December
1979 ECT examination, it was possible to conduct group comparisons.

Table 15. Multiple Prediction of ECf Holistic Score from PWS Characteristics
for White Sample (N=202)
Table 16. Group Comparisons with Respect to Positive, Neutral, and
Negative Perceptions of SpeciiiC Characteristics"

Summary of Group Comparisons

The groups sampled were compared in a number of different ways.

Table 17. Five Most Negatively Perceived Characteristics for Four Groups

OTHER COMPARISONS

In addition to the analyses reported in the preceding sections, several other

comparisons were made during the course of the investigation, as specific issues
of interest arose.

Table 18. Total December 1979 ECT Population Frequencies and

Correlations between Available Variables and ECT Essay Score by Group
Table 19. Essay Writing Performance by Group and SAT-Verbal Score Level

Score Level Comparisons

Since each of the four readers assigned a holistic score in the range of 1 (low) to
4 (high), each essay received a total score in the range of 4 to 16.

Table 20. Essay Writing Performance by Group and TSWE Score Level

E xperienced/lnexperienced Comparisons

Table 22 shows the results of analyses of variance comparing experienced and

inexperienced readers on both the morning reading (Reading I) and the afternoon
reading (Reading II). Recall that Reading I was conducted by assigning a holistic
score and then completing the application form.

Reader Questionnaire Results

Prior to coming to ETS to participate in the reading of essays for the project, each
of the 20 readers was sent a questionnaire (Appendix H).

Table 21. Positive and Negative Perceptions of Essay Characteristics for

Thirteen Score Levels0
 Table 22. Mean Holistic Scores and Characteristic Ratings by
Experienced/Inexperienced Readers
Table 23. Reader Questionnaire Results (Frequencies)

Summary of Other Comparisons

Three additional kinds of comparisons were briefly con sidered: comparisons of

different levels of writing, comparisons of experienced and inexperienced
readers, and comparisons of what English professors say is important with what
emerges as important from analyses of their
judgments. The score level comparisons showed that certain characteristics
operate as influences primarily at very high levels of skill while others operate
primarily at low levels.

SUMMARY AND CONCLUSIONS

The principal objective of the research reported here was to identify

important elements of writing skill as perceived from an examination of a
national sample of writing. The brief, impromptu essays examined represent only
one kind of writing, but an important kind. The first task in the research effort
was to develop a taxonomy of the characteristics of these brief essays through
consultations with writing experts. Twenty characteristics were identified, and
those were organized into three categories: discourse, syntactic, and lexical.
Each characteristic could be perceived positively, negatively, or neutrally.

IMPROVING THE WRITING SKILLS OF AT-RISK STUDENTS THROUGH

THE
 USE OF WRITING ACROSS THE CURRICULUM AND WRITING
PROCESS INSTRUCTION

Theresa A. Dean-Rumsey

Grand Valley State University

ABSTRACT

A plan for improving the writing skills of at-risk students was developed
and implemented- Subjects were fifteen seventh-grade Title I students who
demonstrated weak writing skills. Evidence for the existence o f the problem
included MEAP test scores and teacher observations. A review o f research on
writing pointed to the following solutions to the problem: increasing students
experiences with writing by implementing writing across the curriculum,
improving students knowledge and use of the writing process, and direct
instruction in writing strategies and techniques. Results of the posttest did not
indicate a significant improvement in writing had occurred as a result of the
treatment. However, teacher observations suggested that students writing had
improved in several areas, including students knowledge o f and use of the
writing process.

INTRODUCTION

The writing skills o f many students at Hesperia Middle School need

improvement. The Michigan Educational Assessment Program (MEAP) Writing
Proficiency Test scores for the 1997-98 school year showed that 38.8% o f the
fifthgrade students and 51.1% o f the eighth-grade students were considered
not yet proficient writers. Many o f the non-proficient writers are students who
are considered at-risk o f academic failure due to their low academic
performance, test scores, poverty level, or family situations. Instructional
strategies are needed which will improve at-risk students writing abilities.
Writing skills are an essential component of literacy; students need to be.
proficient writers in order to participate in our literate society. The National
Council of Teachers o f English and the International Reading Association (1996)
have stated that the literacy requirements o f our society are increasing and are
expected to continue to rise. It is estimated that by the year 2020, students will
need powerful literacy abilities to participate fully in society and in the
workplace.
Standardized test scores provide evidence that many students do not have
satisfactory writing skills. While recent curriculum changes have increased
Hesperia Middle School students test scores in mathematics and reading, writing
is an area that still needs improvement.
 Cox, Holden, and Pickett (1997) found that a writing workshop approach
enhanced at-risk students attitudes toward writing. Writing workshop is a natural
learning approach.
Harris and Schaible (1997) stated that writing across the curriculum can
improve students writing and subject knowledge, but only when it is consistently
applied.
Graham & Harris (1997) found that explicit teaching of revising strategies
improved the organization and quality of students papers.
Pope and Beal (1994) reported that a process approach, combined with
guiding and scaffolding students writing experiences, is likely to produce
positive results in writing achievement.
The purpose o f this study is to conduct an experiment to determine
effective strategies for increasing the writing skills o f at-risk students at
Hesperia.A primary goal o f this study is to increase opportunities for students to
write by using writing across the curriculum.This experiment is limited to
studying the effects of instruction in writing strategies and the implementation of
writing across the curriculum on the writing skills of at-risk seventh-grade
students.

In order to build a foundation for the study o f improving the writing skills
of at-risk students, several relevant areas of research will be examined. The
research will focus on the following topics: the importance of improving writing
skills, the instructional needs of at-risk students, writing across the curriculum
programs, using the writing process, effective instructional strategies for
teaching writing, and evaluation.
Writing skills are an essential component of literacy; in order to participate
in the literate society o f the future, students need to be proficient writers.
According to Standards for the English Language Arts, the 1996 report by the
National Council of Teachers of English (NCTE) and the International Reading
Association (IRA), the literacy requirements of our society are increasing and are
expected to continue to rise. Many of the non-proficient writers in the Hesperia
Community Schools are defined as at-risk o f academic failure. According to
Student Services Director Jon Thompson (personal communication September
25, 1998), students are defined as at-risk when they meet one or more o f the
following characteristics: unsatisfactory standardized test scores; failing grades
in core academic subjects; victim of child abuse or neglect; pregnant teenager or
teenage parent; eligible for fi-ee or reduced price lunch due to family poverty;
family history o f school failure, incarceration, or substance abuse; below grade
level performance in English language and communication skills; or atypical
behavior or attendance patterns. Anson (1993) reported that writing across the
curriculum programs seem to have grown from a consensus among educators
that writing is central to learning and should be part of all academic contexts.
Writing across the curriculum proponents believe that teachers of all disciplines
should include writing as part of their course content in order to improve subject-
area learning and improve writing skills. In their article Writing Across the
Curriculum Can Work. Harris and Schaible (1997) stated that anecdotal
evidence
suggests that both students and faculty believe that students improve their
writing and subject-area knowledge in writing across the curriculum-based
courses (p. 31). In her introduction to Writing Across the Curriculum: A Guide to
Developing Programs, Susan McLeod (1992) explained the basic assumptions of
writing across the curriculum programs. One of the main assumptions is that
writing and thinking are closely allied, that learning to write well involves
learning particular discourse conventions, and that, therefore, writing belongs in
the entire curriculum, not just in a course offered by the English department (p.
6). The term writing to leam is often used in interchangeably with writing
across the curriculum. Anne Ruggles Gere (1985), editor o f Roots in the
Sawdust: Writing to Leam Across the Disciplines, made this distinction between
the goals of writing to leam and writing across the curriculum: Writing
across the curriculum aims to improve the quality of writing, while writing to
leam focuses on better thinking and learning (p.5).

The writing process, as defined by the NCTE and IRA (1996), includes the
many aspects o f the complex act o f producing a written communication,
specifically, planning, drafting, revising, editing, and publishing (p.77). In an
article in English Journal. Process Writing and the Secondary School Reality; A
Compromise, Camey (1996) described how she adapted a process writing
instructional approach to fit the curriculum and time constraints o f her teaching
situation. The writing process approach emphasizes student choice in topics and
deadlines, but that is sometimes impossible in schools with a regimented
curriculum. Hillocks (1984) reviewed experimental treatment studies in
composition from 1963-1982 and reported his findings in the article What Works
in Teaching Composition: A Meta-analysis of Experimental Treatment Studies in
the American Joumal o f Education.

To determine the effect of writing across the curriculum and writing

instruction on the writing improvement of at-risk seventh-grade students, a
pretest and posttest based on the Michigan Educational Assessment Program
(MEAP) Writing Proficiency Test will be given.
The sources examined were all valuable in contributing to the knowledge
base for this study. The following sources will be used as a basis for this study:
Collins (1992) on writing across the curriculum instructional methods; Zemelman
and Daniels (1988) for writing to leam instmctional techniques and writing
process philosophy and teaching strategies; and Graham and Harris (1997) for
writing process research and instmctional methods. In the next chapter, the
thesis strategies and methodologies will be described.
COMPONENTS / ACTIVITIES

The purpose of this study was to examine the effectiveness of writing

across the curriculum and writing process instruction on the writing abilities of at
risk seventh grade students. After examining the research on methods to
improve writing skills, two approaches were chosen as a possible solution to this
problem: writing across the curriculum and writing process instruction. Research
has indicated that writing across the curriculum and using a process approach to
writing instruction can contribute to students growth in writing abilities
(Zemelman & Daniels, 1988). The goals for this study were to increase students
writing time, particularly in content-area classes; expand students knowledge
and use o f the writing process; and provide students with direct instruction in
writing strategies across the curriculum. The research design for this study was a
quasi-experimental, single group pretest-posttest design. The experimental
period was September 28 through November 6, 1998. The experimental
treatment included writing across the curriculum, instruction and practice in
using the writing process, and instruction in specific writing strategies across the
curriculum. At the end o f the treatment period, the students were given a
posttest modeled on the MEAP eighth grade Writing Proficiency Test to determine
whether any significant gains had been made in writing skills. The success of the
treatment was measured in two ways; examining the results of the posttest, and
reviewing the teacher observations.

The setting for this study was Hesperia Middle School, located in a small
rural community about sixty miles northwest of Grand Rapids. The middle school,
comprised of grades five through eight, has 406 students. The students come
from predominantly white middle class or poor family backgrounds; 48% of the
middle school students qualify for free or reduced lunch due to family poverty.
Many students move in and out of the district due to family situations. There
were five teachers involved in this study, from the content areas of science,
mathematics, social studies, reading, and English. The teachers all had received
in-service training in using the John Collins method for writing across the
curriculum, and had all used it in their classrooms in varying degrees of
frequency and intensity prior to this study.

Writing is important in all curricular areas. Research has shown that

writing
promotes subject-area learning and critical-thinking skills. There is substantial
evidence that content learning is improved when writing across the curriculum is
used in a rigorous, consistent manner (Harris & Schaible, 1997). Writing is an
effective way to reinforce learning. According to Emig (1977), writing, through
its inherent reinforcing cycle involving hand, eye, and brain makes a uniquely
powerful multidimensional mode for learning (p. 124-125). Writing provides a
way to increase students involvement in lessons, to check for understanding of
concepts, and to stimulate and promote thinking (Collins, 1992).

STRATEGIES AND METHODOLOGIES

The research I did led to the development of the hypothesis for this study;
writing across the curriculum, combined with writing process instruction, will
improve the writing skills o f at-risk seventh-grade students.
The Students
Fifteen seventh-grade students, eight males and seven females, were
selected for this study. These students are all considered to be at-risk of
academic failure due to their unsatisfactory MEAP test scores.

Goals and Objectives

According to Graham and Harris (1997), students cannot develop the skills
for effective writing if they do not write frequently and for extended periods o f
time. Based on researchers recommendations, teachers focused students
attention on two or three specific skills in an assignment (Collins, 1992; and
Zemelman & Daniels, 1988). The skills used as consistent focus correction areas
were as follows; correct paragraph form, with topic sentences and supporting
details; writing complete sentences; and demonstrating use o f all stages o f the
writing process.

Pretest
Since the entire seventh grade would be receiving the experimental
treatment in their content area classes, all students were given the pretest.

Teaching Strategies
Because the teachers participating in the study had received training in
using the Collins (1992) model o f writing across the curriculum, ideas and
strategies from this model were used in developing assignments for the content
area classrooms. Collins categorizes writing assignments into five distinct types,
each with its own purpose and outcomes. Type One is one-draft writing to get
ideas on paper as brainstorming or a prelude to class discussion and is evaluated
for completing the assignment. Type Two writing is a one-draft response to a
teachers prompt that is evaluated for correct content. Type Three writing adds
the component of focus correcting, which directs student and teacher attention
on specific writing or thinking skills in an assignment.

Science Writing Assignments

Students completed seven writing assignments in their science class
during the seven-week experimental period. A variety of writing assignments
were given, and the assignments were structured so that students gradually
moved from writing Type One assignments to writing Type Four assignments.

Mathematics Writing Assignments

Students completed seven writing assignments in mathematics
throughout the study. The writing assignments asked students to create their
own story problems or to explain a process used in problem-solving.

Social Studies Writing Assignments

In their social studies class, students completed seven writing
assignments during the experimental period.
Reading Class Writing Assignments
 Students completed six writing assignments during the experimental
period: four Type One responses, one Type Two response, and one Type Four
paper.

English Class
The focus for English class during the experimental period was providing
more class time for writing and increasing students knowledge and use of the
writing process.

Evaluation
posttest, a practice version o f the MEAP Writing Proficiency Test taken
from the MEAP sCoach Grade 8 Writing (Crowell & Kolba, 1997, p. 77), was given
the week o f November 2 over a three-day period. The format o f the posttest
was the same as described for the pretest. The prompt asked students to write
an editorial for a newspaper explaining why it is important to learn about other
cultures.

ANALYSIS AND CONCLUSIONS

Discussion of Results
A practice version of the MEAP Writing Proficiency Test was administered
as a pretest and posttest to examine the hypothesis that writing across the
curriculum and writing process instruction would improve the writing skills o f at-
risk students.

Conclusions
Evaluation of the pretest and posttest data found no significant differences
in the students scores. The findings did not support the hypothesis that writing
across the curriculum, combined with writing process instruction, would improve
the writing skills of at-risk students.
RECOMMENDATIONS/ PLANS FOR DISSEMINATION

Although this project has been completed, I plan to continue this study
throughout this school year to determine if, with increased time, any
improvement in students writing skills will occur.
 REFERENCES

Anson, C. M. (1993) The future of writing across the curriculum: Consensus and
research. In C. M. Anson, J. E. Schwiebert, & M. M. Williamson (Eds.), Writing
across the curriculum: An annotated bibliography, (pp. xiii-xx). Westport, CT:
Greenwood Press.
Anson, C. M. & Beach, R. (1995). Journals in the classroom: Writing to leam.
Norwood, MA: Christopher-Gordon.
Arkle, S. (1985). Better writers, better thinkers. In A. R. Gere (Ed.), Roots in the
sawdust: Writing to leam across the disciplines (pp. 148-161). Urbana, IL:
National Council of Teachers of English.
Atwell, N. (1987). In the middle: Writing, reading, and leaming with adolescents.
Portsmouth, NH: Heinemann.
Bryson, M. & Scardamalia, M. (1991, March). Teaching writing to students at-risk
for academic failure. In Teaching advanced skills to educationally disadvanted
students. (UD 028 249). (ERIC Document Reproduction Service No. ED 338 725)
Camey, B. (1996). Process writing and the secondary school reality: a
compromise. English Journal. 85 (6), 28-35.
Collins, J. J. (1992). Developing writing and thinking skills across the curriculum:
A practical program for schools. Andover, MA: The Network. 63
Cox, P., Holden, S., & Pickett, T. (1997). Improving student writing skills through
the use of Writing To Leam. (ERIC Document Reproduction Service No. ED 411
520)
Crowell, S. C. & Kolba, E. D. (1997). The MEAP Coach Grade 8 Writing. New York;
Educational Design.
Day, S. L. & Koorland, M. A. (1997, Fall). The future isnt what it used to be:
Student competencies for the 21 century. Contemporarv Education. [On-line], 69.
34-40. Available: First Search
Dougherty, B. J. (1996). The write way: a look at journal writing in first year
algebra. Mathematics Teacher. lOn-linel. 89. 556-60. Available: First Search Emig,
J. (1977). Writing as a mode of leaming. College Composition and
Communication. 28. 122-128.
Fiderer, A. (1986). Math-writing and thinking. In J. Golub (Ed.), Activities to
promote critical thinking (pp. 148-151). Urbana, EL: National Council of Teachers
of English.
Fitzgerald, J. & Markham, L. R. (1987, April). Teaching children about revision in
writing. Paper presented at the annual meeting of the AmericanEducational
Research Association, Washington, D.C. (ERIC Document Reproduction Service
No. ED 282 220)
Forsman, S. (1985). Writing to leam means leaming to think. In A.R. Gere (Ed.),
Roots in the sawdust: Writing to leam across the disciplines (pp. 162-174).
Urbana, IL: National Council of Teachers of English. 64
Fulwiler, T. (1984). How well does writing across the curriculum work? College
English, 46 (2), 113-25.
Gere, A. R. (Ed.). (1985). Roots in the sawdust: Writing to leam across the
disciplines. Urbana, IL: National Council o f Teachers o f English.
Graham, S. & Harris, K.R. (1997). It can be taught, but it does not develop
naturally: myths and realities in writing instruction. School o f Psychology
Review.
[On-line], 26. Available: First Search
Graves, D. Forward. In Atwell, N. (1987) In the middle: Writing, reading, and
leaming with adolescents. Portsmouth, NH: Heinemann. GrifBn, K. (1997).
Writing in the social studies classroom. Voices From the Middle. 4 (21. 31-37.
Harris, D. E. & Schaible, R. (1997). Writing across the curriculum can work.NEA
Higher Education Journal, 13(1), 31-40.
Hillocks, G. (1984). What works in teaching composition: A meta-analysis of
experimental treatment studies. American Journal o f Education. 93 (1), 133-170.
Hillocks, G. (1996). The need for interdisciplinary studies on the teaching of
writing. In Wiley, M.,
Gleason, B., & Phelphs, L.W. (Eds.), Composition in four keys: Inquiring into the
field, (pp. 354-363). Mountain View, CA: Mayfield.
Hodges, V. P. (1993, March). Teaching at-risk students: A quality program in a
small rural high school. Paper presented at the National Conference on Creating
the Quality School, Oklahoma City, OK. (Eric Document Reproduction Service No.
ED 360 131) 65 International Reading Association and National Council o f
Teachers of English. (1996). Standards for the English Language Arts. Urbana, IL:
National Council o f Teachers o f English.
Lehr, J. B. & Harris, H. W. (1988). At-risk, low-achieving students in the
classroom. Washington, D. C.: National Education Association.
McLeod, S. H. (1992). Writing across the curriculum: An introduction. In
S. H. McLeod & M. Soven (Eds.), Writing across the curriculum: A guide to
developing programs (pp. 1-11). Newbury Park, CA: Sage Publications. Michigan
Department o f Education. (1997). Michigan Educational Assessment Program
Handbook. Lansing, MI.
Robertson, J. (1997). Poetry in science. Voices From the Middle. 4 (2\ 7-10.
Pope, C. & Beal, C. Building pathways for at-risk students and their teachers.
Voices From the Middle. 1(2). 3-10.
Sakai, A. & Leggo, C. (1997). Knowing from different angles: Language arts and
science connections. Voices From the Middle. 4 (2), 26-30. Schubert, B. (1987).
Mathematics journals: Fourth grade. In Fulwiler, T.
(Ed.), The Journal book (pp. 348-358). Portsmouth, NH: Heinemann. United
States Department of Education. (1992). Writing Framework for the 1992
National Assessment o f Educational Progress. Washington, D.C. 6 6
Walvoord, B. E., Hunt, L. L., Dowling, H. F. Jr., & McMahon, J. D. (1997). In the long
run: A study of faculty in three writing-across-the-curricuium programs. Urbana,
EL: National Council of Teachers o f English.
Zemelman, S. & Daniels, H. (1988). A community of writers: Teaching writing in
the junior and senior high school. Portsmouth, NH: Heinemann. Zinsser, W. K.
(1988). Writing to leam. New York: Harper & Row.

A STUDY ON IMPROVING THE ENGLISH WRITING SKILLS OF HIGHER

SECONDARY STUDENTS IN MEGHALAYA, INDIA, THROUGH A
THREE-WEEK MODULE.

Evarisha M Syiem

EXECUTIVE SUMMARY
Importance of Communicating through writing in English
Communication through writing is an extremely critical component of
education, livelihood, and basic functionality in a society. Especially, in higher
education where English happens to be the medium of instruction, writing is
extremely important because it is used extensively in communicating with
professors, employers, peers, or just about anyone else.

Problems faced by students using English as a Second Language

However, the ability to write well is not a naturally acquired skill; it is
usually learned or culturally transmitted as a set of practice in formal
instructional settings or other environments. Further, the students skill in writing
in a second language is faced with several challenges. There can be many social
as well as cognitive reasons for this, such as a negative attitude towards the
target language, cultural distance between them and the target language, and
lack of motivation. Students are even ignorant of the basic rules and structural
patterns which they are supposed to have learnt at the school level. Some EFL
teachers may be perplexed by such problems in their writing classes and are
unable to find an efficient way to awaken students imagination and set their
minds working. As a result, students feel that using the English language amidst
all their doubts and uncertainties is a cumbrous affair.
Besides the reasons mentioned above, another major hurdle that students
face is the dearth of time to practise writing in the classroom, with the teacher
available for clarifying their doubts. Coping with the pressure to complete a set
syllabus, teachers hardly have the time to allow students to practise writing.
Writing is a skill and like any other skill, being fluent in it requires plenty of
practice.

Need for a New Paradigm in teaching English writing skills

Teaching English as a second language, especially writing skills, can vary
according to the cultural and academic environment. The situation in North East
India can be quite different from other parts of India due to the high literacy
rates as well as a predominantly Christian population, with its innovative
practices of ritual and rites. Thus, a new paradigm will be needed to explain the
process of developing writing skills in this area, which is the major focus of the
research.
Studies of second language writing are sadly lacking, and little research
done on the L2 writing process among indigenous populations. Students who
have learnt to converse and write in vernacular in their school education, and
rather minimally or not at all in English, would then find communication at the
University level rather difficult and frustrating, unless necessary help is provided
to prepare them properly. This is especially true in Meghalaya, where there are
both external and internal constraints related to the Learner-achievement levels
among tribal students. While there are many courses to enhance the spoken
English and oral communication, very little work is reported on how to improve
the writing skills. There is a great need, therefore, to study the problems faced
by the students and then develop suitable strategies to improve their writing
skills in English, keeping in mind the dearth of time in the academic year.

Research Objectives
This research project explored the development of a short module to
demystify the skill of writing. It allows the students to practise writing alongside
the teachers assistance and guidance. It involves students who, although are
not native users of the English language, yet have learned and used it in speech
through their school years. The aim of this research
was to develop a training course to improve the English writing skills of higher
secondary students, particularly, in Meghalaya. The specific objectives are to
develop a feasible and cost-time-effective training module for enhancing the
writing skills in English; secondly, to evaluate its effectiveness and validate its
applicability; and finally to identify further research strategies on teaching-
learning methodologies in English writing.

Methodology adopted in this Research

Based on selected research articles published on this topic, conceptual
and theoretical frameworks for the research were designed. Modules of varying
duration and intensity were formulated and shared with experienced teachers
and some parents, as well as with a representative sample of high school
students and older students. It was generally agreed that skills in letter writing
should be the first part of the module, followed by writing English compositions.
Necessary training in English vocabulary, and gradual expansion of their grasp of
new vocabulary and essential grammar was also considered essential, to be
instructed in a gradual manner.
With this preparation, research was done on identifying the most appropriate
content as well as method of instruction. Given the academic pressures, it was
finally decided to have three-week course, with the content adjusted accordingly.
The prototype of the course was shared with students, parents and teachers and
the module finalized in terms of hours and timings of instruction, additional home
assignments, and the graded sessions over a period of three weeks. Evaluations
were planned at the end of each week. The tentative training module was
subjected to a pilot study in one school, and analysed in terms of its acceptance,
effectiveness and theoretical soundness, based on which it was modified and
fine-tuned for a proper randomized controlled trial.

Study Design and Sample selected for the Research

A list of higher secondary schools in Shillong was prepared, and these
schools were approached for their cooperation and suitability to participate in the
trial. Based on their consent and suitability, 4 similar schools were chosen, from
which a random allocation was made with 2 schools to test the new module and
the other 2 as matching controls. The study was done in pairs, one experimental
and one control, at a time. The schools were visited and discussions held with
the Administration and concerned teachers. An Orientation programme was
drawn up for the students.

Data Collection, Management and Statistical Analyses

While the experimental schools were administered the newly developed
module, the control schools were given a traditional reading and writing courses
over the period of three weeks. For all the groups mentioned above, the same
procedure was followed as to the sequence of the study. First, a Pretest was
given followed by the three-week training for the Experimental groups and a
normal reading training for the Control groups. At the end of each of the weeks,
posttests were given to assess and evaluate the students performance.
Assessments were made at the end of each week in both the schools using a
standardized system. At the end of three weeks, a Feed-back form was filled by
the students on their comments regarding the course, the positive and negative
aspects, the learning that occurred and their confidence built, as well as their
suggestions for improvement.
Data were entered onto to Microsoft Excel sheets, checked for
completeness and accuracy and monitored weekly. Statistical analysis was done
using SPSS, calculating the differences between pre-course levels and post-
course levels at the end of first week, second week and third week. Paired t-test
was used to determine the statistical significance of the change in Letter writing
and Composition writing skills.
 These changes in the Experimental Group and the Control group were
tested for statistical significance using the t-test and Chi-square test. Frequency
distributions, Means and Standard
Deviations were computed as well as Correlation and Regression coefficients.
Finally, an ANOVA was done on the final scores at the end of three weeks,
between the Experimental and Control groups. Diagrams were drawn to
supplement the tables.

Findings and Conclusions

The Pilot study was conducted with a total of 114 students, in one
representative school. In the randomized controlled trial, there were 114 children
in the experimental group and 128 in the control schools. The average age of the
children was 16 to 17 years.
The findings from the research studies have shown that the new module has
been effective in teaching writing skills, and there was a statistically significant
difference in the performance of students who underwent the new 3-week
programme as compared to controls.
The students expressed satisfaction over the benefits of the training programme,
and found it convenient and useful in the long run. Some students mentioned
that the course did not help to the extent expected, as they had difficulties in
doing the required homework, additional reading and other exercises.
Conclusions were made that it is possible to teach English writing skills for
secondary school level students based on proper pedagogical principles within 3
weeks. The course should include both class-room teaching and home
assignments and the emphasis should be on interactive teaching and active
learning by the students participation. The practice sessions should be logically
arranged from simple to complex, from letter-writing to free compositions. Ample
opportunities should be provided for informal feed-back, corrections and
revisions.
Recommendations
The final training module was made keeping to the duration of three weeks (vide
Annex1). A Training Guide was prepared for the teachers to use along with the
required reading/other training material (Annex 2).
It is recommended that the new module be widely advertised and tested in other
environments to make suitable refinements and adjustments.