Microsoft Word - 6.Tanyer_266-288.docx
1This research is part of the master’s thesis completed at Anadolu University (See Tanyer (2014)).
Tanyer, S. (2017). Domain specific beliefs about writing and
writing performance of preservice English teachers: Is
there any relationship? International Online Journal of
Education and Teaching (IOJET), 4(3), 266-288.
http://iojet.org/index.php/IOJET/article/view/191/172
Received: 28.03.2017
Received in revised form: 30.04.2017
Accepted: 03.05.2017
DOMAIN SPECIFIC BELIEFS ABOUT WRITING AND WRITING
PERFORMANCE OF PRESERVICE ENGLISH TEACHERS: IS THERE ANY
RELATIONSHIP?1
Seray Tanyer
Anadolu University
seraytanyer@anadolu.edu.tr; seraytanyer@gmail.com
Seray Tanyer has been working as a research and teaching assistant since 2011. She received
her BA (2011) and MA (2014) in ELT from Hacettepe University and Anadolu University
respectively. Recently, she has been studying for a PhD in ELT at Anadolu University. Her
research focus areas include foreign language writing, language teacher education and second
language acquisition.
Copyright by Informascope. Material published and so copyrighted may not be published
elsewhere without the written permission of IOJET.
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
266
DOMAIN SPECIFIC BELIEFS ABOUT WRITING AND WRITING
PERFORMANCE OF PRESERVICE ENGLISH TEACHERS: IS THERE ANY
RELATIONSHIP?
Seray Tanyer
seraytanyer@anadolu.edu.tr; seraytanyer@gmail.com
Abstract
Learning as a retrospective phenomenon can make learners transmit their past as an ingredient
while they are (re)structuring their present and future. Previous and present experiences can
form a basis for cognitive, behavioral and motivational factors which can create a cognitive
load for learners and affect their learning process. In this regard, current study aims to
investigate first-year undergraduates’ beliefs about writing and relation of these beliefs to
writing performance in essay writing. A total of 147 students studying in ELT department of a
Turkish university participated in the research. Their domain-specific beliefs about writing
were determined through the Beliefs about Writing Survey (BAWS). Writing performance
was measured on an essay writing task by calculating both overall grade and six component
grades. As a result, multiple regression analysis affirmed that beliefs about writing accounted
for writing performance independently. Pearson correlation values showed that some beliefs
about writing were adaptive and associated with higher writing scores (e.g. “Adapt to the
Audience”). Also, some belief subcategories were associated with each other. The results of
the present study have been discussed along with the related literature on beliefs about writing
and writing performance. Implications/suggestions related to the coursework, writing
practices and future research have been presented.
Keywords: beliefs about writing, foreign language writing, writing performance, preservice
English teachers, teacher training.
1. Introduction
Writing, the neglected skill of foreign language education, “today is not a frill for the few,
but an essential skill for the many” (National Commission on Writing, 2003, p. 11). We
require this skill to flourish as a student, as an employer and as a citizen because foreign
language writing is one of the critical components of academic foundation that students need,
an essential prerequisite in the workplace and a critical requirement to be active citizens of a
globalizing world. Therefore, this skill becomes valuable for different groups including
preservice teachers of English who are potential teachers of all other groups. This assumption
has motivated the researcher to explore quite a lot about writing discipline, and its first and
earlier practices in undergraduate composition classes. Following this perspective, domain-
specific beliefs about writing have been chosen as the target focus in this study.
Social cognitive theory of Albert Bandura refers to the significant role of beliefs in human
learning and performance (Bandura, 1989). Self-efficacy beliefs are one of them which are
mainly related to a person’s belief in oneself to perform tasks and reach particular goals (see
Bandura (1997) for further reading). More than thirty years, research on participants from
primary school children to university students have already reinforced the relationship
between writing self-efficacy, writing apprehension and writing performance (e.g., McCarthy,
Meier, & Rinderer, 1985; Meier, McCharty, & Schmeck, 1984; Pajares & Johnson, 1994;
Pajares & Valiante, 1999; Prat-Sala & Redford, 2012; Tanyer, 2015). However, only recently,
Tanyer
267
the social cognitive view of writing has been extended via the concept of beliefs about writing
which has been related to writing performance as well as writing self-efficacy and
apprehension (i.e., Sanders-Reio, 2010; Sanders-Reio, Alexander, Reio, & Newman, 2014).
This belief category (terminologically domain specific beliefs about writing) has been built on
studies from the fields of educational psychology, writing and rhetoric, and concentrates on
beliefs about what good writing is, what god writers do as well as the elements related to
writing process itself.
The exploration of preservice English teachers’ beliefs might be crucial in terms of various
reasons. Firstly, it is possible for them to transmit their past as an element to learning
environments of today and future. Therefore, if preservice teachers get promoted to reflect
their own beliefs, they can construct or reconstruct their belief systems about writing.
Moreover, beliefs about writing can be related to learners’ writing performance as shown in
the literature (e.g., Perry, 2011; Sanders-Reio, 2010; Sanders-Reio et al., 2014; White &
Bruning, 2005). Also, like other foreign language skills such as speaking and listening,
writing can be regarded as one of the neglected skills until university education. Probably
because of this fact, this skill has been observed as a challenge for first-year preservice
English teachers, and it would be worthy to discover what kinds of beliefs about writing are
held by this specific group. In addition, while organizing a writing course in an EFL teacher-
training program, there may be a need to learn future teachers’ beliefs about a specific domain
that they would teach in the future which is “writing” in this study. Identifying their personal
beliefs and the additional details about these beliefs in depth by means of valid and reliable
instruments might present valuable implications for teaching writing skill in a teacher
education program.
1.1. Literature review
This review presents a conceptual framework for the exploration of research on beliefs
about writing as well as a summary and discussion of the studies conducted in this area. The
topics related to the research on writing beliefs have been ranging from innateness of writing
skill to role of audience, mechanical and substantive writing skills, specific models of writing
beliefs as well as transmissional, transactional and domain specific beliefs about writing. The
first topic of empirical studies about writing beliefs concentrates on the role of giftedness in
writing ability. With 247 undergraduates, Palmquist and Young (1992) examined the
relationship between the beliefs in the innateness of writing ability and four other variables
that were writing apprehension, self-assessment of writing skills, the confidence in mastering
writing skills and genres, and previous experience with writing teachers. According to the
results, the belief in the innateness of writing skill might “contribute to these students’
apprehension about writing” (Palmquist and Young, 1992, p.151). Also, the participants
believing that writing skill was innate-gift tended to be less confident in their ability to
become professional writers. The undergraduates carrying this belief also reported their
experiences with their previous writing teachers less favorably. Therefore, Palmquist &
Young (1992) concluded that the belief in innateness might provoke unprompted limitations
about undergraduate courses and future careers which required writing skills.
From the same point of view, Charney, Newman & Palmquist (1995) added one different
variable that was writing performance, and investigated the relationship between beliefs about
the innateness of writing ability, students’ assessments about their own writing, writing
apprehension and writing performance of 446 undergraduate students. The findings revealed
that participants believing that writing could be learned were more likely to enjoy writing
although they did not accordingly score higher on their writing assignments. Also,
participants enjoying writing more tended to label themselves as good writers, and the female
participants were prone to regard writing skill as something that could be learned. In addition,
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
268
they enjoyed writing more, and scored higher in writing assignments than the male ones. As
suggested by the two studies above, the belief in giftedness might have prejudicial effects on
student writers and trigger limitations in writing career (Palmquist & Young, 1992).
Therefore, it would be valuable to question this belief in that writing lecturers may need to
become aware of it and combat its negative effects.
Another perspective in beliefs about writing research deals with the role of audience in
writing process. The analysis of audience is an important element of planning process because
it decides the format, the language, the information included, and even the use of figures and
graphics. In this regard, Nelson (2008) indicated the role and importance of audience and
addressed three issues that were “writing related to reading, writer related to reader, and text
related to text” regarding them as central to written discourse (p. 547). Her study revealed that
the participants considered readers and writers interacting in a bilateral aim instead of isolated
existences. It was indicated that the products of writers varied according to the different types
of audience by means of an adaptation process. Nelson also noted the evidence of
developmental differences between younger and older students.
The four other researchers, Miller and Chorney (2008) and Beach and Friedrich (2006),
focused on three main dimensions of writing that are persuasion, audience and argument, and
discussed how writers adapted the organization, content, tone of their arguments according to
target audience, their age, their assumed attitudes, and to the specific discourse community.
Also, emphasizing the significance and pervasiveness of writing in the workplace, Beaufort
(2008) researched how writers in workplace adapted their voice, tone, level of clarity, and
stated that the writers had decided on their choice of words in response to the power groups.
As claimed by Miller and Charney (2008), the notion of influencing and persuading readers
has been emphasized since Aristotle. However, in school environment, the written products
that students produce may not address a real aim or a real audience. For this reason, it can be
valuable to question students’ beliefs about this inauthenticity and the role of audience in a
foreign language environment.
Mechanical and substantive skills are two other components of writing skill which address
different issues: while mechanical skills are interested in grammar, spelling, punctuation and
style; substantive skills attend to organization, development, clarity, and cohesion. The
differences between these two skill types can be observable in research. In their study of how
students’ beliefs about writing affect the product and writing process, Graham, Schwartz &
MacArthur (1993) made a distinction between mechanical and substantive skills. Via open-
ended interviews, the 4th, 5th,7th and 8th grade students with and without learning disabilities
were asked about their beliefs and knowledge with respect to what good writing was, what
good writers did, why some students had trouble in writing, and how they would write a paper
for a younger child. The participants were also asked to evaluate a text written by a child by
employing their knowledge and beliefs about writing. The findings revealed that the better
writers who were also older and normally achieving students tended to highlight substantive
skills instead of mechanical skills in their definitions of good writing. In their accounts of who
good writers were and what they did, these participants emphasized writing processes over
written product. To sum up, Graham, Schwartz & MacArthur (1993) claimed that “The
knowledge, attitudes, and beliefs that students hold about writing play an important part in
determining how the composing process is carried out and what the eventual shape of the
written product will be” (p.246). As writing teachers, if we uncover the students’ beliefs about
mechanical and substantive issues, these beliefs might be (re)shaped by writing instruction
that they would receive.
Some scholars have also started to develop models on beliefs about writing combining
different variables. The earlier empirical study of beliefs about writing has been published by
Tanyer
269
Silva and Nicholls (1993) who designed their scales based on six traditions of discourse
theory. Based on these traditions, Silva and Nicholls included some goals and beliefs in two
different scales (i.e., “Writing Goals” and “Beliefs about the Causes of Success in Writing”).
They also applied three more scales that were “Intrinsic Commitment to Writing”, “Dualism
Scale” and “Perceived Ability Scale”. The results revealed that students with beliefs referring
to substantive issues liked writing more than the ones holding beliefs stressing “Surface
Correctness and Form”. Silva & Nicholls (1993) also stated that beliefs about writing might
reflect writing teachers’ styles and classroom culture. Therefore, as in the Bandura’s model,
the effect of environment on the person could be observed in writing classrooms, as well.
In addition to those above, Ellen Lavelle has published a number of studies about students’
approaches to writing (e.g., Lavelle, 1993, 2001, 2003; Lavelle & Guarino, 2003; Lavelle,
Smith & O’Ryan, 2002; Lavelle & Zuercher, 2001). She started her research by developing a
questionnaire, The Inventory of Processes in College Composition (IPCC; Lavelle, 1993), a
factor analysis of which has provided five different writing approaches of college students:
Elaborationist Approach, Low Self-Efficacy Approach, Reflective-Revisionist Approach,
Impulsive Approach and Procedural Approach. Those five approaches were collected under
two broad categories as deep and surface approaches (Lavelle & Guarino, 2003). Writers
taking a deep approach are more meta-cognitive, more involved in their writing and regard
themselves as a real operator in meaning making. These writers tell a strong sense of
audience, carry holistic views of writing tasks, and they are directed more toward meaning of
the written product than form. For them, revision is an important part of writing process, and
they tend to revise and reflect thoroughly on their product. Writers taking the surface
approach are less aware of writing process and audience, and they are less dedicated to their
written product. Writing is not a learning source for them; they are more rule-bound, focus on
mechanical errors in writing, and instead of revising, they edit their work at the surface level.
As well as the models mentioned above, White and Bruning (2005) adapted the earlier
works of Schraw and Bruning (1996, 1999) to writing which investigated transmissional and
transactional beliefs about reading. According to the new model, writers with high
transmissional beliefs see writing principally as a channel of transmitting authoritative
knowledge to readers with minimum addition of writer’s own perspective while writers with
high transactional beliefs view writing as a medium to combine what they have learned with
their own knowledge and perspectives. In order to measure transmissional and transactional
beliefs about writing, White and Bruning (2005) developed the Writing Beliefs Inventory and
examined the relations among 170 undergraduates’ beliefs about writing, writing self-
efficacy, writing apprehension, past writing experiences and writing performance. The results
indicated that beliefs about writing were related to writing performance because the
participants with transmissional beliefs had significantly lower writing scores while those
with transactional beliefs had higher writing scores. Students with high transmissional scores
also had less affective and cognitive engagement with writing and were less likely to write for
pleasure. On the other hand, students with high transactional scores spent more time for
writing and were more likely to find writing pleasurable.
More recently, Bruning, Dempsey, Kauffman & Zumbrunn (2011) extended the work of
White and Bruning (2005) by revising Writing Beliefs Inventory and surveyed 556 eleventh
graders from two high schools to investigate the relationship among implicit beliefs about
writing, affects towards writing, writing self-efficacy, writing grades and statewide writing
assessment scores as well as English/Language Arts course enrollment. Results demonstrated
that transactional beliefs were significantly related to liking writing, self-efficacy for writing
ideation, self-efficacy for writing conventions, self-efficacy for writing self-regulation, self-
reported grades, and the statewide writing assessment score. Also, students in more advanced
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
270
classes of English/Language Arts course had higher transactional beliefs and lower
transmissional beliefs. Following that study, Perry (2011) investigated 153 college students’
implicit beliefs about a specific writing task and associations of those beliefs to writing score
with the measures of “Writing Habits and Beliefs Scale”, “Writing Beliefs Inventory-
Revised” (Bruning et al., 2011), “Liking Writing Scale” and “Beliefs about Intelligence”
scale. Results revealed that college students held implicit beliefs about the specific writing
task and those beliefs were related to liking writing and beliefs about intelligence. However,
transmissional and transactional beliefs did not affect scores of writing task. All in all, this
study indicated that learners approached “writing with a unique set of beliefs, assumptions,
and motivations”, and they entered “the classroom with a wide variety of skill sets,
experiences, and prior knowledge” (Perry, 2011, p.96).
Referring to a scarcity, Sanders-Reio (2010) indicated that while investigation on writing
self-efficacy beliefs and its relation to writing apprehension and writing performance had
started to become accumulated, the research area of beliefs about writing was limited.
Therefore, she examined the association between domain specific beliefs about writing,
writing self-efficacy and writing apprehension, and their relations to writing performance on a
three-phase study. The first two phases were assigned to instrument construction and
validation while the last phase investigated the relations among the target variables. The
participants were 207 pre-service teachers studying in the College of Education, and the
measures were “Beliefs about Writing Survey”, the modified “Writing Self-Efficacy Scale”
(Zimmerman and Bandura, 1994), the modified “Writing Apprehension Test” (Daly and
Miller, 1975), and a demographic survey. Writing performance of the preservice teachers was
assessed via a structured five-page paper written for the educational psychology course.
The results revealed that four of the beliefs about writing – Expert Orientation, Writing
Supports Thinking, Address Substantive Issues First and Mechanical Errors Are Shameful -
appeared to be adaptive in that they positively correlated with all or some of the grades for
writing performance or with other adaptive beliefs. Expert Orientation also correlated
positively with writing self-efficacy and enjoyment for writing while Writing Support
Thinking had the highest correlation to enjoyment of writing. Hierarchical regression analysis
revealed that beliefs about writing independently explained 12% of the variance in writing
performance. Apprehension about making grammatical and other mechanical errors had a
strong negative effect on writing performance. Lastly, after controlling for domain specific
writing beliefs, writing self-efficacy weakly predicted writing performance.
A following study of Sanders-Reio et al. (2014) also followed Kellogg’s (2008) cognitive
model of writing development and investigated the relations among beliefs about writing,
writing self-efficacy, writing apprehension and writing performance. “The Beliefs About
Writing Survey”, “the Writing Self-Efficacy Index” and the modified “Writing Apprehension
Test” were administered to a total of 738 undergraduates, and writing performance was
evaluated based on a class paper. According to the findings, beliefs about writing accounted
for writing scores significantly and, while the beliefs, Audience Orientation and Recursive
Process, were the positive predictors of the scores, Transmission and Transaction were the
negative predictors. As for the other variables, the writing self-efficacy positively and
apprehension about grammar negatively predicted writing performance while writing
apprehension as a block was not a significant predictor.
More recently, Tanyer & Subaşı (2016) conducted a qualitative study and investigated 26
preservice English teachers’ beliefs about EFL writing via interviews. In the study, the
participants’ beliefs about good writing and writers, writing ability and the factors affecting
their beliefs about good writing were interrogated. As a result, it was found that participants
approached good writing and writer characteristics and the nature of writing ability with
Tanyer
271
particular belief sets. The scholars also interpreted their findings according to the writing
scores gathered by the participants. Based on these interpretations, they claimed that the
relationship between writing beliefs and writing performance had been reinforced by their
study.
All in all, the findings above support the possibility that writing beliefs can be an
influential variable while teaching writing skill. Following the research above dealing with
domain-specific beliefs about writing, the present study has two main purposes. Firstly, it
aims at investigating first year undergraduates’ domain-specific beliefs about writing.
Secondly, it questions the role of these beliefs in undergraduates’ writing performance. In line
with these purposes, it has been assumed that this study would provide a profile of preservice
English teachers by reporting their beliefs about what good writing is, what good writers do in
addition to the writing processes, writing tasks, writing skills and the procedures these tasks
and skills involve. The research questions guiding the study are as follows:
RQ1. Do preservice English teachers hold any domain specific beliefs about writing?
RQ2. Is there any relationship between domain specific beliefs about writing and writing
performance?
RQ3. Do domain specific beliefs about writing predict writing performance?
2. Methodology
2.1. Participants
The participants included 147 first-year preservice teachers studying in the ELT
Department of a Turkish university. In this department, students must take two main writing
courses in their first year (i.e. Written Communication and Academic Writing and Report
Writing). In this study, the ones from all eight sections of “Academic Writing and Report
Writing Course” participated. Each section of that course comprised almost 30 students;
however, the ones that had failed in previous years and retook the course were excluded from
the analysis. Most of the participants were female with the proportion of 74.8% (N=110),
while 25.2% of them were male (N=37). Additionally, the average age of them was 19.69
(SD= 2.12).
There are several reasons for the selection of this population of interest in this research.
Firstly, they receive considerable amount of practice and instruction in writing through two
semesters, so learning more about them and undergraduate writing might facilitate the
development of writing instruction for this sample. The second reason is that the participation
of this sample would facilitate comparison with and the extension of much of the existing
research (e.g., Bruning et. al., 2011; Sanders-Reio, 2010; Sanders-Reio et al., 2014; White &
Bruning, 2005) about pre-service teachers’ domain specific beliefs about writing.
2.2. Instruments
In the present study, two data sources were combined which were a recent survey of
beliefs about writing and writing scores. The survey administered and measurement of writing
performance have been described below.
2.2.1. Survey
The participants indicated their domain-specific beliefs about foreign language writing on
a five-point likert scale which was The Beliefs about Writing Scale (BAWS; Sanders-Reio,
2010). BAWS had specifically been designed with Hispanic first-year preservice teachers
who were enrolled in College of Education in south Florida, USA. The original BAWS
comprised 76 items with 14 subscales, although a four subscales-version of it have recently
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
272
been developed and used by Sanders-Reio et al. (2014). In the current study, the first version
with 76 items and 14 subscales were administered.
Beyond the beliefs about foreign language writing, the survey battery also asked
participants to provide relevant background information about themselves such as their age,
gender and year of study. The respondents were also required to reveal some more details
such as their attitudes toward foreign language writing and their past educational experiences.
Some items of the instrument theorized to be vague for the first-year preservice teachers were
disambiguated with their synonyms or with some examples. Moreover, three writing
instructors all of whom were experts in ELT and lecturing in ELT Department reviewed the
survey battery and approved the modifications. The comparison between the initial and
modified version of the items have been listed in Table 1:
Table 1. The original and modified version of five items of BAWS
Original Modified
9. It’s important to develop a distinctive
writing style.
19. Writers need to immerse themselves in
their writing.
39. Papers with typos are terrible
embarrassment.
45. During revision, one should carefully
check one’s manuscripts for both
substantive and mechanical problems.
46. Good writers demonstrate their skills at
crafting complex sentences.
69. It’s humiliating to give a PowerPoint
presentation with typos and misspellings.
9. It’s important to develop a distinctive
(peculiar, original) writing style.
19. Writers need to immerse themselves in
(involve deeply in) their writing.
39. Papers with typos (misspellings) are
terrible embarrassment.
45. During revision, one should carefully
check one’s manuscripts for both
substantive (organization, development
etc.) and mechanical (grammar, spelling
etc.) problems.
46. Good writers demonstrate their skills at
crafting (creating) complex sentences.
69. It’s humiliating to give an essay with
typos and misspellings.
2.2.2. Writing Performance
Writing performance was assessed via the grades participant received on the papers they
wrote for the Academic Writing and Report Writing Course. During an in-class exam, the
students were provided with three optional writing prompts and were free to choose any of
them for essay writing. Also, based on their topic, they were supposed to decide the genre
type which they would write in using APA citation techniques appropriately. The papers were
assessed analytically via ESL Composition Profile of Jacobs, Hartfield, Hughey and Vormuth
(1981) which includes five main rows: Content, Organization, Vocabulary, Language Use,
and Mechanics. As Andrade & Boulay (2003) have argued, using such profile for assessment
can support learning and development of writing skills by laying out clear, concrete
characteristics of good writing.
2.3. Data Collection and Analysis Procedures
2.3.1. The Survey
At the beginning of spring semester, randomly selected thirty-two (32) preservice teachers
studying in ELT Department firstly indicated their domain specific beliefs about foreign
language writing via BAWS. The purpose of applying this instrument to that small group was
to discover whether the items were clear and definite for the respondents. As no negative
feedback was received and there had been no additional modification on the survey, the data
Tanyer
273
gathered from this quite small sample were combined with the main study. Thus, a total of
147 participants answered the survey battery in a class hour.
The ELT Department demands their candidates to be at B2 level according to the standard
language levels of Common European Framework of Reference for Languages (CEFR) while
accepting them to the department. This level was assumed to be appropriate for preservice
teachers to comprehend the items of the BAWS by the faculty members. For this reason, the
original English version of the survey was applied during the data collection process.
Furthermore, the preservice teachers were demanded to approve their voluntarily participation
to complete the survey package by signing a consent form. In that consent form, the
participants were enlightened about the purpose of the research and asked for their
permission.
2.3.2. Writing Performance
As for the writing performance, the participants’ first midterm exam scores were taken into
consideration. During an in-class exam, the preservice teachers wrote their own essays. In the
exam paper, the students were provided with three optional writing prompts and were free to
choose any of them. Also, based on their topic, they were supposed to decide the genre type
they would write.
2.3.3. Data Analysis
A number of statistical calculations have been performed in order to address the research
questions. For RQ1 (i.e., Do preservice English teachers hold any domain specific beliefs
about writing?), descriptive statistics such as means and standard deviations were used. For
RQ2 (i.e. What is the relation between beliefs about writing and writing performance?), the
Pearson correlations were computed and analyzed between the independent (beliefs about
writing) and dependent (writing performance) variables. To answer RQ3 (i.e. Do domain
specific beliefs about writing predict writing performance?), a standard multiple regression
analysis was employed to determine the unique variance explained by beliefs about writing in
writing performance.
3. Results
In the following headings, reliability of measures, descriptive statistics, and the findings of
correlational and standard multiple regression analyses have been presented.
3.1. Reliability of measures
To assure that the Beliefs about Writing Scale (BAWS) was a reliable measure, internal
consistency of measures was used by computing the Cronbach’s alphas for the total scale and
all subscales (see Table 2). The Cronbach’s alpha for the total scale was .855, which was
close to the entire value of original BAWS (i.e. 0.87; Sanders-Reio, 2010, p. 116). The
Cronbach’s alphas for the fourteen subscales of BAWS ranged from .524 to .797, which had
ranged from .61 to .80 in the original scale. As eliminating any items did not provide higher
reliability, none of the items had been removed from the scale.
Table 2. Reliability coefficients for BAWS and its subscales
N of items Cronbach’s α
1. Transmissional 6 .569
2. Writing Supports Thinking 4 .797
3. Writing Is a Personal and Emotional Experience 6 .653
4. Writing Is an Innate Gift 5 .762
5. Basics (Mechanics) First 4 .578
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
274
6. Address Substantive Issues First 5 .524
7. Writing Is an Iterative Process 8 .597
8. Minimize Revision 7 .604
9. Write to Impress 4 .609
10. Use Plain English 6 .638
11. Adapt to the Audience 8 .664
12. Clarity Is Essential 3 .532
13. Development Is Important 5 .589
14. Mechanical Errors Are Shameful 5 .778
Beliefs about Writing Survey (TOTAL) 76 .855
Writing performance was assessed via the scores participants received on the essays they
wrote for the first midterm exam on three optional topics. As for the reliability of writing
scores, two graders each of whom were experienced instructors of writing and had been
lecturing for at least fifteen years evaluated the participants’ papers following the dimensions
of ESL Composition Profile (Jacobs et. al., 1981). The mean of the two graders’ scores was
used in the study. Based on these two score sets, a correlational analysis was operated in order
to calculate the inter-rater reliability as instructed by Gay (1992). The correlation value
between the total scores given by the two scorers was .94.
3.2. Descriptive statistics
The values below demonstrate descriptive statistics about the participants’ profile, past
educational and writing experiences in addition to means and standard deviations of the
subscales.
Figure 1. Do you like writing in English? (N=147)
The first question, Do you like writing in English?, inquired participants’ attitudes toward
foreign language writing. As revealed by Figure 1, most of the students (77.6%) seem to have
developed a positive attitude toward writing in English, while the rest 22.4% of them reported
that they did not like writing in English.
The remaining open-ended questions provided us some more details about the participants’
past writing practices in English. For example, the hours of English courses they took in a
week during high school ranged from 2 to 14 hours with a mean of 10.71. But, when the
course hours allocated for foreign language writing was searched, it was indicated that 67.3%
of the participants (N=99) had not done any practices of writing in English at high school (see
Figure 2 below). The time allocated for writing practice for the rest of the participants was
also quite limited varying between one (1) and five (5) hours with a decreasing proportion
from 17.7% to 0.7% respectively.
77,6%
22,4%
Yes
No
Tanyer
275
Figure 2. How many hours of English courses in a week were allocated to improve your
writing skills at high school? (N=147).
In Table 3, the means and standard deviations of the fourteen (14) subscales in BAWS
were demonstrated. The subscales were ordered in a descending order from the one with the
highest average to the one with the lowest average.
Table 3. Means and standard deviations of the subscales in BAWS (N=147)
M SD
1. Development Is Important 4,125 0,424
2. Clarity Is Essential 4,063 0,571
3. Writing Supports Thinking 4,056 0,663
4. Adapt to the Audience 3,996 0,437
5. Writing Is an Iterative Process 3,938 0,427
6. Writing Is a Personal and Emotional Experience 3,818 0,524
7. Address Substantive Issues First 3,693 0,574
8. Writing Is an Innate Gift 3,449 0,805
9. Basics (Mechanics) First 3,449 0,635
10. Use Plain English 3,353 0,554
11. Write to Impress 3,258 0,727
12. Transmissional 2,997 0,584
13. Mechanical Errors Are Shameful 2,851 0,793
14. Minimize Revision 2,409 0,524
As presented in Table 3, the beliefs that were Development Is Important, Clarity Is
Essential and Writing Supports Thinking had the highest means while the beliefs,
Transmissional, Mechanical Errors Are Shameful and Minimize Revision, had the lowest
mean scores. The other eight belief categories about writing were in between these two outlier
sets. Descriptive statistics showed that the upper three beliefs were highly agreed by the
participants; on the contrary, the last three beliefs seemed to be disagreed.
In Table 4 (below), the results of one-sample t-test that had compared the averages of each
subscale with the middle value (i.e. 3) were presented. Among the subscales which were
agreed, the highest average belonged to the belief, Development Is Important (M= 4.125),
while the lowest average belonged to the belief that writers should Write To Impress the
audience (M= 3,258). Although the lowest average had a mean of 3,258, this value was
higher than the middle value (i.e., 3) with a t-value of 4.307 and a probability value of .000.
67,3%
17,7%
9,5%
2,7% 2,0% 0,7%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Never 1 Hour 2 Hours 3 Hours 4 Hours 5 Hours
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
276
Therefore, it was found that the participants agreed all the first eleven (11) beliefs about
writing listed below.
Table 4. Summaries of one-sample t-test comparing the averages of variables with the middle
value of the likert scale (i.e. 3) (N=147)
Variable Mean SD t df p <
1. Development Is Important 4,125 0,424 32.120 146 .000
2. Clarity Is Essential 4,063 0,571 22.575 146 .000
3. Writing Supports Thinking 4,056 0,663 19.288 146 .000
4. Adapt to the Audience 3,996 0,437 27.639 146 .000
5. Writing Is an Iterative Process 3,938 0,427 26.632 146 .000
6. Writing Is a Personal and Emotional Experience 3,818 0,524 18.910 146 .000
7. Address Substantive Issues First 3,693 0,574 14.656 146 .000
8. Writing Is an Innate Gift 3,449 0,805 6.757 146 .000
9. Basics (Mechanics) First 3,449 0,635 8.564 146 .000
10. Use Plain English 3,353 0,554 7.731 146 .000
11. Write to Impress 3,258 0,727 4.307 146 .000
12. Transmissional 2,997 0,584 -.047 146 .963
13. Mechanical Errors Are Shameful 2,851 0,793 -2.265 146 .025
14. Minimize Revision 2,409 0,524 -13.668 146 .000
On the other hand, the last two beliefs were significanly disagreed which were Mechanical
Errors Are Shameful (M=2.851) and writers should Minimize Revision (M=2.409) whose
means were lower than the middle value (i.e. 3) with t-values of (-2.265) and (-13.668)
respectively. Also, they had probability values of (.025) and (.000) which were lower than
.05. Therefore, it was apparent that respondents disagreed with these beliefs. The last belief,
Transmissional (M=2.997), was not agreed or disagreed by the subjects. In other words, they
were unsure about this belief since the mean of it was so close to the middle value of the scale
(i.e. 3). Additionally, it had a very low t-value (-.047) and the probability value of it (i.e.,
.963) was not lower than .05. In this regard, the pre-service English teachers participating in
the present study were found to be unsure about the Transmissional belief category.
3.3. Correlations
In order to answer Research Question 2 (i.e. Is there any relationship between beliefs about
writing and writing performance?), Pearson correlations among dependent variables (i.e. the
subscales of the BAWS), and the Pearson correlations between the dependent variables and
the independent variable (i.e. writing performance) were computed.
3.3.1. Correlations Among the Subscales of the BAWS
Firstly, the correlational values among the dependent variables have been reported, and the
correlation matrix in Table 5 signified that a number of subscales had been statistically
significantly correlated among each other and with writing performance. Some previous
studies have theorized (e.g., White & Bruning, 2005; Sanders-Reio, 2010) that beliefs about
writing can be adaptive since “they either reflect expert practice, support writing process,
and/or tend to be as associated with better grades on writing assignment” (p.151). The belief
categories theorized to be adaptive were: 1) Adapt to the Audience, 2) Clarity Is Essential, 3)
Development Is Important, 4) Writing Is An Iterative Process, 5) Use Plain English, 6)
Substantive Issues First, 7) Writing Supports Thinking and 8) Writing Is A Personal and
Emotional Experience.
Tanyer
277
In the current study, Adapt To The Audience significantly and positively correlated with all
of the theorized adaptive beliefs. This means that preservice teachers believing that writing
should be adapted to the audience were more likely to believe that Development Is Important
(r =.56, p < .01), Clarity Is Essential (r =.54, p < .01), Writing Is An Iterative Process (r
=.47, p < .01), Writing Supports Thinking (r =.31, p < .01), foreign language writers should
Use Plain English (r =.29, p < .01), Writing Is A Personal and Emotional Experience (r =.26,
p < .01) and that writers should Address Substantive Issues First (r =.24, p < .01). To the
contrary, first-year preservice teachers subscribing to the belief, Adapt To The Audience, were
more likely to hold three of the maladaptive beliefs which were Basics (Mechanics) First (r
=.303, p < .01), Transmissional (r =.24, p < .01), and Write to Impress (r =.20, p < .01).
Second adaptive belief correlating significantly and positively with all theorized adaptive
beliefs is Development Is Important. Therefore, those believing that writers should explain
their thoughts effectively were more likely to believe that Clarity Is Essential (r =.62, p <
.01), writers should Adapt To The Audience (r =.56, p < .01), Writing Is An Iterative Process
(r =.50, p < .01), writers should Use Plain English (r =.37, p < .01), Writing Is A Personal
and Emotional Experience (r =.34, p < .01), Writing Supports Thinking (r =.31, p < .01), and
writers should Address Substantive Issues First (r =.28, p < .01). In contrast, participants
subscribing to this belief (i.e., Development Is Important) were more likely to believe that
writers should master Basics (Mechanics) First (r =.38, p < .01), and should Write to Impress
(r =.18, p < .01) the reader.
The third writing belief subcategory theorized to be adaptive is Clarity Is Essential.
According to the Table 5, the respondents believing that foreign language writers should
convey information clearly were more inclined to believe that Development Is Important (r
=.56, p < .01), writers should Adapt To The Audience (r =.54, p < .01), Writing Supports
Thinking (r =.39, p < .01), Writing Is An Iterative Process (r =.39, p < .01), writers should
Use Plain English (r =.34, p < .01), Writing Is A Personal and Emotional Experience (r =.31,
p < .01), and that writers should Address Substantive Issues First (r =.28, p < .01). On the
contrary, the subjects believing that Clarity Is Essential were more likely to believe that
writers should master Basics (Mechanics) First (r =.30, p < .01) and should transmit
authorities’ words into their writing (Transmissional, r =.23, p < .01).
Five beliefs about writing that are Transmissional, Write to Impress, Minimize Revision,
Writing Is An Innate Gift and Basics (Mechanics) First are theorized to be maladaptive for
they contradict with “expert writing practice” and have a tendency to link to lower scores on
writing tasks (Sanders-Reio, 2010; p. 154). In the current study, three of these maladaptive
beliefs, Transmissional, Write to Impress and Basics (Mechanics) First, correlated with other
three maladaptive beliefs. The first one, Transmissional, positively and significantly
correlated with Write to Impress (r =.30, p < .01), Minimize Revision (r =.23, p < .01) and
Basics (Mechanics) First (r =.18, p < .05). This means that participants who regarded writing
as a means of transmitting scholarly information to the audience with almost no contribution
also believed that writers should Write To Impress and Minimize Revision while writing.
Minimize Revision supports the notion that writers write it appropriately in the first time
and need no revision, and stands as a counterpoint to the concept of expert orientation,
especially to the belief that Writing Is An Iterative Process. This belief correlated positively
and significantly with the other beliefs theorized to be maladaptive including Transmissional
(r =.23, p < .01), Write To Impress (r =.20, p < .05) and Writing Is An Innate Gift (r =.19, p
< .05). As anticipated, it also significantly and negatively associated with Writing Is an
Iterative Process (r =.19, p < .05). The belief, Write to Impress, which suggested that writers
should use big words and attract their readers, was also correlated with the three beliefs.
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
278
Table 5. Intercorrelations among the subscales of beliefs about writing survey and writing performance
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1. Transmissional --
2. Writing Supports
Thinking .090 --
3. Writing is a Personal
Emotional Experience .216
** .265** --
4. Writing is an Innate Gift .107 -.164* .136 --
5. Basics (Mechanics) First .186* .130 .069 .156 --
6. Address Substantive
Issues First -.016 .132 .244
** .047 -.208* --
7. Writing is an Iterative
Process .188
* .378** .282** -.057 .300** .137 --
8. Minimize Revision .232** -.131 .122 .194* .048 .220** -.192* --
9. Write to Impress .303** .128 .325** .214** .118 .081 .152 .205* --
10. Use Plain English .234** -.023 .102 .214** .160 .204* .273** .077 -.069 --
11. Adapt to the Audience .244** .313** .268** .051 .303** .245** .475** -.025 .209* .299** --
12. Clarity is Essential .233** .394** .314** .055 .309** .284** .391** -.071 .147 .349** .543** --
13. Development Is
Important .159 .315
** .349** .100 .389** .282** .501** -.077 .189
* .377** .568** .626** --
14. Mechanical Errors are
Shameful .159 -.018 -.011 .174
* .391** -.234** .000 .022 .217
** .106 .099 .032 .068 --
15. Overall Grade -.085 .016 .053 -.135 -.022 .153 .170* -.128 .026 .107 .269
** .136 .171* .118
** Correlation is significant at p < .01. *Correlation is significant at p < .05.
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
279
theorized to be maladaptive that were Transmissional (r =.30, p < .01), Writing Is An Innate
Gift (r =.21, p < .01) and Minimize Revision (r =.20, p < .05).
The other maladaptive belief, Writing Is An Innate Gift, was also positively and
significantly correlated with Minimize Revision (r =.19, p < .05) and Write To Impress (r
=.21, p < .01). In contrast, this belief also negatively associated with the belief of Writing
Supports Thinking (r =.16, p < .05) which proposes that writing can help writers better
understand what they think. An interesting result was that Basics (Mechanics) First
correlated positively with only one maladaptive belief which was Transmissional (r =.18, p
< .05) while correlating positively with four adaptive beliefs, in which the correlational
values ranged from .30 to .38 (p < .01).
The last belief, Mechanical Errors Are Shameful, was not theorized as adaptive or
maladaptive. In our study this belief correlated positively and significantly with three
maladaptive beliefs including Basics (Mechanics) First (r =.39, p < .01), Write To Impress (r
=.21, p < .01) and Writing Is An Innate Gift (r =.17, p < .05). It also negatively and
significantly associated with Address Substantive Issues First (r =.23, p < .01) which had
been theorized as an adaptive belief in the literature.
3.3.2. Correlations Between the Independent Variables and Writing Performance
One of the characteristics of adaptive beliefs is that they tend to match with higher scores
on writing tasks (Sanders-Reio, 2010). As seen in Table 5, the belief that writers should
Adapt To The Audience was significantly and positively correlated with overall writing
performance (r=.26, p< .01), which proposes that those who put emphasis on audience-
orientation were more probable to receive higher grades on their written work. In addition to
the audience adaptation, the belief, Development Is Important, positively and significantly
correlated with the overall grade (r=.17, p< .05). This result suggested that those believing
that writers should present logical and convincing arguments were more likely to score higher
just like the ones believing that audience-adaptation was significant. Lastly, the belief,
Writing Is An Iterative Process, correlated positively with the overall score (r=.17, p < .05).
These results indicated that students subscribing to the beliefs held by expert writers such as
audience-orientation, presenting logical and convincing arguments as well as the ones
viewing writing as a process of reviewing were inclined to have higher scores on their papers.
On the other hand, although some beliefs such as Minimize Revision (r= -.12, NS),
Transmissional (r=-.85, NS) and Basics Mechanics First (r= -.02, NS) negatively correlated
with writing performance, these correlational values were quite low and not significant.
In sum, the exploration of the correlations between the beliefs about writing and overall
writing performance justifies the view that some beliefs are prone to be adaptive while others
are tendentious to be maladaptive although the correlational values were not statistically
significant in our findings regarding the maladaptive beliefs.
3.4. Standard Multiple Regression
Standard multiple regression analysis was conducted to decide the variance in writing
performance explained by domain specific beliefs about writing totally and individually. The
prediction model summarized in Table 6 (below) was statistically significant
F(14,132)=2.192, p < .05 and accounted for approximately 18.9% of variance of writing
score (R2 = .189, Adjusted R2 = .103). This value is fairly higher in comparison with Sanders-
Reio’s (2010) outcomes in which all beliefs about writing explained 11.8% of the variance
(p<.001).
Tanyer
280
Table 6. Standard multiple regression analysis results
Model b SE-b Beta
Pearson r
(β) sr
2 Structure
Coefficient
Constant 41.167 14.484
1. Transmissional -3.331 2.016 -.150 -.085 .016 -.195
2. Writing Supports Thinking -2.761 1.829 -.141 .016 .013 .036
3. Writing Personal and Emotional .465 2.276 .019 .053 .000 .122
4. Writing Is an Innate Gift * -2.843 1.403 -.176 -.135 .025 -.311
5. Basics (Mechanics) First -2.738 2.114 -.134 -.022 .010 -.050
6. Substantive Issues First 2.630 2.213 .116 .153 .008 .352
7. Writing Is an Iterative Process 2.596 3.155 .085 .170 .004 .391
8. Minimize Revision -2.160 2.294 -.087 -.128 .005 -.294
9. Write to Impress .493 1.677 .028 .026 .000 .059
10. Use Plain English .977 2.263 .042 .107 .001 .246
11. Mechanical Errors Are
Shameful* 3.489 1.492 .213 .118 .033 .271
12. Development Is Important .847 3.674 .028 .171 .000 .394
13. Clarity Is Essential .336 2.578 .015 .136 .000 .313
14. Adapt to the Audience* 7.741 3.126 .260 .269 .037 .619
According to Table 6, writing scores were primarily predicted by the belief category of
Adapt to The Audience (β=.26, p<.05), and to a lesser extent by two other subcategories, that
were Mechanical Errors Are Shameful (β=.21, p<.05) and Writing Is An Innate Gift (β=-.17,
p<.05). Also, the raw and standardized regression coefficients of the predictors together with
their correlations with the writing performance, their squared semi-partial correlations and
their structure coefficients were shown in the table. The belief category, Adapt To The
Audience, received the strongest weight in the model followed by Mechanical Errors Are
Shameful and Writing Is An Innate Gift. Overall, the model indicated that higher Adapt To
The Audience and Mechanical Errors Are Shameful scores predicted higher writing grades
while the higher Writing Is An Innate Gift scores predicted lower writing grades in our
research sample.
4. Discussion
This current research aimed to identify preservice English teachers’ domain specific
beliefs about writing and examine the relations between various writing beliefs and writing
performance. The research literature on writing beliefs does not go far away in time.
Moreover, it is limited and related to different samples and disciplines such as reading
research (e.g., White & Burning, 2005), writing pedagogy (e.g., Silva & Nichols, 1993), early
childhood and/or elementary teacher education (e.g., Graham, Schwartz & MacArthur, 1993).
Only a recent study of Sanders-Reio (2010) focused specifically on writing processes and
practices of expert writer candidates. Following her, the current study attempted to contribute
to the literature on domain specific beliefs about writing applying the BAWS to the
preservice English teachers.
One characteristic of adaptive beliefs was that they tended to match with higher scores on
writing tasks (Sanders-Reio, 2010). According to our findings, the belief that writers should
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
281
Adapt To The Audience was significantly and positively correlated with overall writing
performance. This proposed that those who put emphasis on audience-orientation were more
probable to receive higher grades on their written work. In addition to the audience
adaptation, the belief that Development Is Important positively and significantly correlated
with the overall writing grade. This result suggested that the students who believed that
writers should explain their thoughts and feelings effectively were more likely to score higher
as were those who believed that audience-adaptation was crucial and had a significant role in
writing process. Lastly, the belief, Writing Is an Iterative Process, correlated positively with
the overall score. These results indicated that the students subscribing to the beliefs held by
expert writers such as audience-orientation, effective essay development as well as the ones
viewing writing as a process of reviewing were inclined to have higher scores on their papers.
These findings also support Sanders-Reio’s (2010) and Sanders-Reio et al.’s (2014)
outcomes in which audience-adaptation, attaching importance to development and viewing
writing as an iterative process were the three of the beliefs held by expert writers and
contributed positively to the writing performance. On the other hand, although some beliefs
such as Minimize Revision, Transmissional and Basics (Mechanics) First negatively
correlated with writing performance in our outcomes, these correlational values were not
significant and were quite low. Two of these beliefs that were Minimize Revision and
Transmissional had been negatively and significantly correlated with writing performance
and found maladaptive in previous studies as well (e.g. Sanders-Reio, 2010; Sanders-Reio et
al., 2015; White & Bruning, 2005). This means that some beliefs about writing tend to be the
negative predictors of writing performance even in different research contexts.
Nevertheless, the findings of the study did not support one of the outcomes of White and
Bruning (2005). The Transactional belief that had originated from the research on reading
and claimed that writers were supposed to be interested in their writing both emotionally and
cognitively were divided into two different subscales in Sanders-Reio’s (2010) newly
developed BAWS. These divided subscales were Writing Supports Thinking and Writing Is A
Personal and Emotional Experience. Despite effecting writing performance positively and
significantly in the research of White and Bruning (2005), both beliefs did not have any
significant effect on writing performance in the current study. This can mean that our
students do not regard writing as a mirror on which they can watch and evaluate their own
ideas and/or they do not need to immerse themselves deeply and/or develop a distinctive
writing style to become a good writer. One of the underlying reasons for this finding can be
preservice teachers’ short-term interaction with foreign language writing. However, our
results confirmed the findings of Burning et. al. (2011) in which both Transmissional and
Transactional beliefs did not affect the scores on writing, and did not support the findings of
Sanders-Reio et al.’s (2014) in which Transmission and Transaction was the significant
negative predictors of writing performance.
As researchers and teacher trainers, we are pleased to find out that the mean scores of
preservice teachers in the ELT department were quite high for the three adaptive beliefs. The
first one, Development Is Important, has the highest mean score among all others which is
followed by Adapt To The Audience and Writing Is An Iterative Process in the fourth and
fifth order respectively (See Table 3). So, it can be said that the first-year preservice teachers
tend to implement the requirements of becoming an expert writer. As it is known that
majority of the participants (i.e., 67.3%, see Figure 2) have never practiced foreign language
writing, the writing classes and its components at university will be the first environment in
which the teacher candidates can shape their beliefs about foreign language writing.
Therefore, as teacher trainers, we should benefit from these beliefs of students as much as
possible.
Tanyer
282
According to the standard multiple regression, all beliefs about writing accounted for
approximately 19% of variance of writing score. This value is higher in comparison with
Sanders-Reio’s (2010) and Sanders-Reio et al.’s (2014) outcomes in which the all beliefs
about writing explained 11.8% and 8.4% of the variance in writing scores respectively.
Writing score was primarily predicted by the belief category that was Adapt To The
Audience, and to a lesser extent by two other subcategories, that were Mechanical Errors Are
Shameful and Writing Is An Innate Gift. On one level, the outcome of present study supports
the findings reported in Sanders-Reio (2010) and Sanders-Reio et al. (2014) since the belief
category, Adapt To The Audience, was positive predictor of writing performance in those two
studies. The findings of the study also revealed that the most adaptive variable in this
research was the belief, Adapt To The Audience, because it was the only belief category that
both positively correlated with writing grades and explained statistically significant amount
of variance in the writing scores. It was also positively correlated with other variables related
to good writing that were Development Is Important and Writing Is an Iterative Process,
which strengthens the level of adaptiveness of it. The other supporting point for this
adaptiveness was that this belief also positively and significantly correlated with Mechanical
Errors Are Shameful, which is one of the adaptive beliefs according to our findings.
There is one more issue that needs to be discussed related to these quantitative data. As
stated above, Mechanical Errors Are Shameful is one of the adaptive beliefs since this belief
explained the variance in the writing scores. As proposed by Sanders-Reio (2010), finding
this belief as adaptive is quite unexpected since “shame is a negative emotion” (p. 201). It has
been argued that the preservice teachers’ capacity of using writing mechanics might be an
effective factor for this issue. For instance, this belief “may motivate students with moderate
mechanical skills to” overuse these skills “not to be shamed”, or it may cause preservice
teachers having weak mechanical skills stay away from “writing and facing the shame
associated with” this writing component (p.201). Besides, this belief might associate with
higher grades of the participants due to their writing instructors’ evaluation and scoring
criteria. From a different point of view, regarding mechanical errors as shameful might have
resulted in dealing with this issue successfully without no excuse for the preservice teachers
because most of our participants did not strongly believe that Mechanical Errors Are
Shameful (M=2.851) with a t-value of -2.265 although it significantly and positively
contributed to the total writing score. Also, some writing mechanics such as punctuation rules
can be believed to be a requirement for good foreign language writing, which may shed light
on why the belief that Mechanical Errors Are Shameful explained approximately 2% of the
variance in writing score. When participants were asked what they did with their draft before
it was subscribed to their teachers (see Tanyer, 2014; 2017), they stated that they checked
both their grammar and other writing mechanics such as typos, spelling and punctuation. All
these findings may be one of the explanations of why the belief Mechanical Errors are
Shameful had been found as an adaptive belief and contributed positively to writing scores in
the present study.
As a result of the analysis of data from the survey, the belief, Adapt To The Audience, was
found as an adaptive belief. This finding was also supported with the interview results in
Tanyer & Subaşı (2016). In that study, the nineteen percent (19%) of preservice teachers
believed that good foreign language writing aimed to address and impress audience. The
same beliefs were also discovered to be one of good writer characteristics because, according
to the interviewees, good writers should be aware of their audience (46%). In the same study,
also, the belief that Development Is Important was also confirmed both as a good property of
foreign language writing (e.g., Developing good and creative ideas, 58%) and good foreign
language writer (e.g., Producer of knowledgeable and convincing essays, 27%) with the
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
283
interview results. Moreover, the other adaptive belief, Writing Is An Iterative Process, also
associated with the features of good writing and good writers by the participants. Therefore,
it can be argued that the preservice teachers have started to discover and internalize expert
writing orientations, which is a desirable and promising finding.
As stated above, two other beliefs, Mechanical Errors Are Shameful and Writing Is An
Innate Gift, explained writing performance individually according to our outcomes. The first
one (i.e., Mechanical Errors are Shameful) was found adaptive, and this result was also
supported with the interviews conducted in Tanyer & Subaşı (2016). 8% of the participants
stated that good foreign language writing was required to be “accompanied with punctuations
rules”, and all those respondents (100%) were high-achieving students. Moreover, 58% of the
participants expressed that complex and correct sentences must be formed with advance
grammar in good foreign language writing, and 67% of that group had also received higher
scores in their writing exams. This means that interview results of Tanyer & Subaşı (2016)
are in line with the survey results in terms of adaptiveness degree of the belief, Mechanical
Errors Are Shameful.
The last belief, Writing Is An Innate Gift, was one of the maladaptive beliefs in the
literature (e.g. Palmquist & Young, 1992; Charney, Newman & Palmquist, 1995; Sanders-
Reio, 2010). According to our research sample, that belief had the capacity to negatively
explain writing performance. This means that in line with the previous studies, the students in
our sample who tended to view writing as an innate gift were likely to score lower in their
writing exams, as well. This finding has also been supported with the interview results of
Tanyer & Subaşı (2016). As for writing ability, 42% of teacher candidates viewed writing
both as an innate talent and a skill that can be improved with appropriate instruction, teacher
feedback and student effort. However, 39% of participants characterized writing skill only as
an innate gift, which means that it is almost impossible to become a good writer no matter
how a novice writer puts effort to success. Lastly, only 19% of teacher candidates viewed
writing as a skill that could be developed via enough training and student attempt.
Interestingly, 80% of the respondents who had regarded foreign language writing as an innate
gift were low-achieving students while 60% of the interviewees who had viewed it as an
improvable skill were high-achieving students. Moreover, 73% of the participants who
regarded foreign language writing both as an innate gift and an improvable skill (42%) had
received higher scores in their exams as well. All these interview findings have confirmed the
maladaptive tendency of the belief, Writing Is An Innate Gift, for our sample, and they might
be the explanation of why this belief was found as maladaptive in our current research
environment.
5. Conclusion
Social cognitive theory anticipates that beliefs about writing are associated with success
and failure. Within the scope of this theory, self-efficacy in writing and its association with
apprehension and success in writing has been discussed in a number of research studies;
however, research on domain specific beliefs about writing and its relation to writing
performance has been limited. Therefore, the main purpose of the current study was to
examine first-year preservice English teachers’ beliefs about writing and the relation of these
beliefs to writing performance in essay writing.
As for the relationship between beliefs about writing and writing performance, the
findings indicated that the beliefs, Adapt To The Audience, Development Is Important and
Writing Is An Iterative Process, were significantly and positively correlated with overall
writing performance. According to this finding, those who put emphasis on audience-
orientation, those who believe that writers should explain their thoughts and feelings
Tanyer
284
effectively and those who view writing as a process of reviewing and revising were more
probable to receive higher grades on their written work. On the other hand, although three
beliefs, Minimize Revision, Transmissional and Basics Mechanics First, negatively correlated
with writing performance, these correlational values were not significant and were quite low.
Overall, the first three beliefs mentioned above (i.e., Adapt To The Audience, Development Is
Important and Writing Is An Iterative Process) were prone to be adaptive associating
positively with writing performance while the others (i.e., Minimize Revision, Transmissional
and Basics Mechanics First) were tendentious to become maladaptive although statistical
values regarding the maladaptive ones were not significant in our sample.
As for the domain-specific beliefs about writing predicting writing performance, the
findings revealed that all beliefs about writing as a block accounted for approximately 19%
of variance of writing score (p<.05). Writing performance was primarily predicted by the
belief category of Adapt to The Audience, and to a lesser extent by two other subcategories
that were Mechanical Errors Are Shameful and Writing Is An Innate Gift. While two of these
beliefs (i.e., Adapt to The Audience and Mechanical Errors Are Shameful) accounted for
approximately 3% and 2% of variance in total writing scores respectively, the belief, Writing
Is An Innate Gift, explained 2% of variance in lower writing scores. Therefore, the findings
indicated that higher Adapt to The Audience and Mechanical Errors Are Shameful belief
scores predicted higher overall writing grades while the higher scores of the belief, Writing Is
An Innate Gift, predicted the lower writing scores in our research environment.
5.1. Pedagogical Implications
The present study highlights the existence of preservice English teachers’ domain specific
beliefs about writing, and it concludes that domain-specific beliefs about writing are one of
the influential factors of writing discipline and writing performance. As a training assistant, I
could observe that “addressing course participants’ beliefs about writing” would provide
another road to “writing competence and to more positive and productive attitudes” toward
this discipline (Sanders-Reio, 2010, p. 219). The results of this study indicated that all
domain specific beliefs about writing explained 19% variance in writing grades. For this
reason, in addition to personal classroom observations, the scholarly findings attribute value
to writing beliefs while training preservice English teachers on how to write. That is why the
probable implications that would be stated based on these findings can be essential for
teacher educators.
Firstly, writing instructors can donate their undergraduate writers with particular strategies
such as how to predict and respond to the questions of target audience and adapt their
message to them because there are some writing components that need to be adapted
according to the audience such as the format taken, the information included, and the
language and graphics used. Furthermore, first-year novice writers may be trained for
effective peer or individual revision and editing techniques that they can apply before
submitting their drafts. As stated in the results, the beliefs associated with higher writing
grades were mostly related to “expert writing guidelines and practices” (Sanders-Reio, 2010,
p. 212). Moreover, one of the beliefs related to expert writing, Writing Is An Iterative
Process, which advocates writing as a process of editing and revising, was discovered to be
associated with high writing grades in our study. Based on this finding, it can be claimed that
writing instructors should both observe, research and identify expert writers’ practices both in
academic and authentic contexts, and furnish their learners with these expert writer
qualifications. It should be remembered that as confirmed by the study of Tanyer & Subaşı
(2016; i.e., factors and people effecting beliefs about good writing), school environment and
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
285
its components (e.g., teachers, curriculum, feedback etc.) are the potential factors shaping
beliefs about good writing.
All in all, the findings imply that beliefs about writing can be taken as a crucial leverage
point for teaching writing. For this reason, writing instruction can be modified to “emphasize
the mindsets and approaches associated with adaptive beliefs and minimize those related to
maladaptive and ineffective beliefs.” (Sanders-Reio et al., 2014; p. 10). One way to achieve
this can be giving homework or doing assignments which will inspire learners to focus on
their readers’ characteristics and interests. Writing instruction can also help learners to
present logical and convincing arguments because the belief, Development Is Important, has
been labelled adaptive in nature by the participants. They can be presented with strategies to
explain their opinions and findings effectively, to produce and convey qualified ideas, and to
become logical and convincing of their own views.
5.2. Limitations
There are some points that need to be characterized as limited in the current study. Firstly,
the Beliefs about Writing Survey adopted had originally been developed for the first-year
preservice teachers in a second language environment although it was applied to the
preservice English teachers in a foreign language context in Turkey. However, beliefs about
writing can be specific to culture, writing instruction provided and learning context.
Therefore, students’ beliefs can change based on these factors, or some other beliefs about
writing can exist. To overcome these limitations, it is possible to replicate the study with
different populations, or to discover other possible beliefs about writing and causal relations
of them. Moreover, despite being persistent with some earlier research studies (e.g., Sanders-
Reio, 2010; Sanders-Reio et al., 2014; Shell, Murphy, & Bruning, 1989), the methodology
used to decide students’ writing performance has not reflected the possible variance in the
long-term performance because only one grade received on an in-class exam paper was
included in the study.
5.3. Suggestions for Further Research
The findings of this research may call the need for more examination of the variables by
means of larger and more representative samples in Turkey. Therefore, some beliefs can be
found as more or less adaptive or maladaptive in different learning environments of Turkish
universities, or for different writing tasks. Moreover, possible effect of instructional and
assessment procedures and writing expertise on beliefs about writing can be investigated via
longitudinal studies. Finally, it has been theorized that “there is also a cognitive link mediated
by the writer’s choice of strategies or a student’s openness to instruction in specific
strategies” (Sanders-Reio, 2014; p. 10). For instance, learners may pay attention to
instruction on how to revise only if they believe that successful writers also revise. Therefore,
the possible relationship between beliefs about writing and readiness to instruction, and the
effect of this relationship on writing performance can be investigated as a follow-up research
study.
Tanyer
286
References
Andrade, H. G., & Boulay, B. A. (2003). Role of rubric-referenced self-assessment in
learning to write. The Journal of Educational Research, 97(1), 21-34.
Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44(9),
1175-1184.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman.
Beach, R., & Friedrich, T. (2006). Response to writing. In C. A. MacArthur, S. Graham, & J.
Fitzgerald (Eds.), Handbook of writing research (pp. 222-234). New York: Guilford.
Bruning, R., Dempsey, M., Kauffman, D., & Zumbrunn, S. (2011). Writing viewed as
transmission and transaction: An empirical study of implicit writing beliefs. Poster
session presented at the annual meeting of the American Educational Research
Association, New Orleans.
Charney, D., Newman, J. H., & Palmquist, M. (1995). “I’m just no good at writing”:
Epistemological style and attitudes toward writing. Written Communication, 12(3),
298-329.
Cramer, R. L. (2001). Creative power: The nature and nurture of children’s writing. New
York: Logman.
Daly, J. A., & Miller, M. D. (1975). The empirical development of an instrument to measure
writing apprehension. Research in the Teaching of English, 9(3), 242-249.
Gay, L. R. (1992). Educational research: Competencies for analysis and application (4th
ed.). New York: Merrill.
Graham, S., Schwartz, S. S., & MacArthur, C. A. (1993). Knowledge of writing and the
composing process, attitude toward writing, and self-efficacy for students with and
without learning disabilities. Journal of Learning Disabilities, 26(4), 237-249.
Jacobs, H. L., Hartfield, V. F., Hughey, J. B., & Vormuth, D. R. (1981). Testing ESL
composition: A practical approach. Rowley, Mass: Newbury House Publishers.
Kellogg, R. T. (2008). Training writing skills: a cognitive developmental perspective.
Journal of Writing Research, 1(1), 1-26.
Lavelle, E. (1993). Development and validation of an inventory to assess to processes in
college composition. British Journal of Educational Psychology, 63(3), 489-499.
Lavelle, E. (2001). Writing styles of college students. Journal of College Reading and
Learning, 32(1), 60-67.
Lavelle, E. (2003). Writing approaches of college students: A relational perspective. TLAR,
8(2), 5-17.
Lavelle, E., & Guarino, A. J. (2003). A multidimensional approach to understanding college
writing. Educational Psychology, 23(3), 295-305.
Lavelle, E., Smith, J., & O’Ryan, L. (2002). The writing approaches of secondary students.
British Journal of Educational Psychology, 72(3), 399-418.
Lavelle, E., & Zuercher, N. (2001). The writing approaches of university students. Higher
Education, 42(3), 373-391.
McCarthy, P., Meier, S., & Rinderer, R. (1985). Self-efficacy and writing: A different view
of self-evaluation. College Composition and Communication, 36(4), 465-471.
International Online Journal of Education and Teaching (IOJET) 2017, 4(3), 266-288.
287
Meier, S., McCarthy, P. R., & Schmeck, R. R. (1984). Validity of self-efficacy as a predictor
of writing performance. Cognitive Therapy and Research, 8(2), 107-120.
Miller, C. R., & Charney, D. (2008). Persuasion, audience, and argument. In C. Bazerman
(Ed.), Handbook of research on writing (pp. 583-598). New York: Erlbaum.
National Commission on Writing. (2003, April). The neglected “R”: The need for a writing
revolution. Retrieved November 2, 2013, from
http://www.writingcommission.org/report.html
Nelson, N. (2008). The reading-writing nexus in discourse research. In C. Bazerman (Ed.),
Handbook of research on writing (pp. 435-450). New York: Erlbaum.
Pajares, F., & Valiante, G. (1999). Grade level and gender differences in the writing self-
beliefs of middle school students. Contemporary Educational Psychology, 24(4), 390-
405.
Palmquist, M., & Young, R. (1992). The notion of giftedness and student expectations about
writing. Written Communication, 9(1), 137-168.
Perry, K. R. (2011). Implicit beliefs about writing: a task-specific study of implicit beliefs.
(Doctoral Dissertation). Retrieved from http://digitalcommons.unl.edu/cehsdiss/119/.
Prat-Sala, M. & Redford P. (2012). Writing essays: does self-efficacy matter? The
relationship between self-efficacy in reading and in writing and undergraduate students’
performance in essay writing. Educational Psychology, 32(1), 9–20.
Sanders-Reio, J. (2010). Investigation of the relations between domain-specific beliefs about
writing, writing self-efficacy, writing apprehension, and writing performance in
undergraduates. (Master’s Thesis). Retrieved from ProQuest Digital Dissertations.
(AAT 3409651).
Sanders-Reio, J., Alexander, P., A., Reio, T., G. & Newman, I. (2014). Do students’ beliefs
about writing relate to their writing self-efficacy, apprehension, and performance?
Learning & Instruction, 33, 1-11.
Schraw, G., & Bruning, R. (1996). Readers’ implicit models of reading. Reading Research
Quarterly, 31(3), 290-305.
Schraw, G., & Bruning, R. (1999). How implicit models of reading affect motivation to read
and reading engagement. Scientific Studies of Reading, 3(3), 281-302.
Shell, D. F., Murphy, C.C., & Bruning, R. H. (1989). Self-efficacy and outcome expectancy
mechanisms in reading and writing achievement. Journal of Educational Psychology,
81(1), 91-100.
Silva, T., & Nicholls, J. G. (1993). College students as writing theorists: Goals and beliefs
about the causes of success. Contemporary Educational Psychology, 18(3), 281-293.
Tanyer, S. (2014). The relationship between preservice English teachers' domain specific
beliefs about writing and writing performance: A sample study. (Master’s Thesis).
Avaliable from https://tez.yok.gov.tr/UlusalTezMerkezi/tezSorguSonucYeni.jsp.
(Thesis No: 375298).
Tanyer, S., (2015). The role of writing and reading self-efficacy in first-year preservice EFL
teachers’ writing performance. Procedia-Social and Behavioral Sciences, 199, 38-43.
Tanyer, S., & Subaşı, G. (2016). Domain specific beliefs about writing: A qualitative inquiry
on first-year pre-service English teachers. ELT Research Journal, 5(4), 278-303.
Tanyer
288
Tanyer, S. (2017). How do first-year pre-service English teachers revise their written works?
The reported practices of revision process. Manuscript submitted for publication.
White, M. J., & Bruning, R. (2005). Implicit writing beliefs and their relation to writing
quality. Contemporary Educational Psychology, 30(2), 166-189.
Zimmerman, B. J., & Bandura, A. (1994). Impact of self-regulatory influences on writing
course attainment. American Educational Research Journal, 31(4), 845-862.