Factors affecting writing

achievement: mapping
teacher beliefs
Associate Professor Claire Wyatt-Smith
Director, Centre for Applied Language, Literacy & Communication
Studies, Griffith University, Queensland, Australia

Dr Geraldine Castleton
Head, Continuing Professional Development, Institute of Education,
Worcester University College

Abstract
The intersection of teacher beliefs with writing achievement in
schooling is a key concern of this paper. The paper reports part of a
two-year Australian study that set out to examine in detail how it is that
teachers judge Year 5 students' literacy achievement using writing as
the case instance. In what follows, we examine the data in the form of
concept maps that the teachers themselves made available showing
their beliefs about, and insights into the factors that affect student
writing achievement. Drawing on these maps, we highlight the range
of teacher-identified factors, including those relating to in-class
behaviour, motivation, attitudes to school learning, social and cultural
backgrounds, oracy and even life circumstances. Additionally, w e
address how the identified factors function, operating either as stand-
alone elements or within a dynamic network of inter-relationships.
~~

Key words
Teacher beliefs, writing achievement, factors affecting writing
achievement

Introduction

In this paper we explore teacher beliefs about students' writing

achievement in schooIing . Using teache r-generat ed representa t io ns of
the various factors that teachers self-report as affecting student writing
achievement, the paper makes available insights into teachers' beliefs

0 Nate and Contributors 2004 39

Factors affectina writina achievement

about such achievement, and opens the discussion of the various subject
positions (Kress, 1985) available to teachers as they do their work of
teaching and assessing classroom writing. By taking an interest in the
positions that teachers occupy on writing achievement, we are centrally
concerned with the ways ofknowing (Belenky et al, 1986) that the
teachers have available to them, and also the ways of being, believing, and
valuing that they routinely call upon in their interactions with student
writing and student writers.

The starting point for the paper is our interest in what Katz (2000) referred
to as 'knowledge as arising in the subjective activities of particular
subjects' (p.141). Working from this position, we assume that teachers,
quite properly, occupy a special place in their roles as mediators between
the child and the formal curriculum. From this vantage point, w e argue
that the act of making implicit beliefs explicit is a necessaryfirst step in
considering the relationship between the expectations of formal
curriculum and teacher expectations, especially those that have potential
to impact on teachers' notions of writing achievement.

Three related questions are taken up in the paper: First, what do Year 5
teachers identify as the factors affecting the quality of students' writing?
Second, how do the teachers represent those variables, including any
relationships among them and their relative influence on quality? Finally,
what is the value of making the opening for teachers to share their
conceptions of writing achievement, extending to the pedagogical and
assessment implications of such conceptions?

Backgroundto the study

To set the scene for what follows, we provide some background

information relating to the context in which our investigation has occurred.
The data drawn on in the paper were collected as part of an Australian
Research Council funded two-year study that aimed to provide a
quantitative and qualitative picture of the role of human judgment in
educational assessment. In the study, teachers' assessments of literacy
achievement were examined as instances of judgment in action, the focus
being on the inter-relationships among the various operative facets of
students' performance. This approach makes it possible to examine and
directly gauge the impact of evaluative frameworks that constitute the
official, publicly available assessment agenda as well as the contribution of
otherwise implicit but nonetheless significant determinants of teachers'
judgments. Readers interested in published reports of the study are
advised to see Wyatt-Smith, Castleton, Freebody and Cooksey, 2003.

Thirty-seven teachers from twenty-one schools participated in the study,

the teachers representing a range of approximately two to twenty-five

40 English in Education Vo1.38No.1 2004

 Claire Wyatt-Smitha n d Geraldine Castleton

years teaching experience in state primary schooling. The selection of

schools included cohorts representing a range of socio-economic as well
as cultural and linguistic backgrounds and was drawn mostly from
metropolitan and surrounding areas of a large Queensland city. The
teachers met in groups of four or five, for a full day of data collection,
involving five sessions. In some groups there were two teachers from a
single school, but wherever possible, this was avoided, preferring instead
to enable teachers to engage with the day's activities as individuals rather
than as teaching pairs or teams.

Of primary interest in this paper is the first session in which we invited the
teachers to work individually in developing a concept map of those factors
that they took to affect students' writing achievement. To further illustrate
the salience of information in these maps, relevant segments of teacher
talk from session two are drawn on where appropriate. In this session,
each teacher was paired with a research assistant to tape record 'a think-
aloud' session during which the teacher was encouraged to speak out the
factors he/she was thinking about while marking their own student's
scripts. In this task, the teachers' own class samples are referred to as 'in
context' samples, as they know the students and the pedagogy involved in
the production of the writing samples. Readers interested in more detail
about the sessions of the data collection day are advised to see Appendix
A.

Concept mapping is a procedure applied in a diverse range of research and

practice settings, including science education, teacher education,
cognitive science and management science, to generate individual
conceptual frameworks based on specific items (Kinchin & Hay, 2000).
First devised by Novak (1990, 1998)as a means for improving science
teaching and learning, concept maps provide a useful metacognitive tool
for promoting understanding of individuals' cognitive structures. The
structure of a map is taken to be unique to its author, reflecting his/her
experiences, beliefs and biases as well as hidher understanding of
concepts or topics and relationships between them. The construction of a
concept map illustrates two fundamental properties of understanding; the
representation and the organization of ideas (Halford, 1993). For our
purposes, concept mapping was useful not only in bringing to light the
factors that the teachers took to be affecting writing achievement, but also
as a means of generzting insights into how they represented these. This
included how the teachers represented the factors, one in relation to the
other, and also how they showed the strength of each factor in relation to
the set.

Procedurally, we gave each teacher an A3 sheet of paper with the words

'Writing Achievement' boxed in the centre of the page and a pad of post-it
notes. Initially, the teachers were asked to write on the note pages those

0 Nate and Contrfbutors 2004 41

Factors affecting wrfting achievement

factors that they considered affected writing achievement. We advised

them not to attend to the order in which they recorded the factors, the aim
being to capture each one as it occurred to them. In effect, the writing on
the notes was a type of stream-of-consciousness activity, where the
ordering of the factors was not a first priority. The teachers worked
individually and did not discuss their selection of factors with other
participants. Then we asked the teachers to place each separate post-it
with its identified factor on the A3 sheet, using arrows to indicate the
relative strength of each factor, on a scale of 1 to 3 (3 representing a strong
link to writing achievement; 2, a moderate link; and 1, a weak link).

As the final step in the process, we asked the teachers to draw in

association links across the factors, marking links with single or bi-
directional arrows. At the end of the day, the teachers could revisit their
individual maps for the purpose of confirming or amending the factors that
they identified earlier in the day. As part of the study design, we decided
not to share the maps among the participating teachers on any given day
or across days, the emphasis being on encouraging teachers to bring to
the surface beliefs and insights that they may not disclose if required to
share in such a group setting. The maps included in this paper are
authentic teacher-ge nerate d materia Is , being actu a I outcomes of the
process described above. We provide more detailed discussion of specific
maps later in the paper.

Teacher-identifiedfactors

We undertook the analysis of the maps in two phases, with Phase 1

focusing on the range and diversity of factors that the teachers identified,
and Phase 2, focusing on what the maps revealed about how individual
teachers drew on and combined the factors, networking them in
distinctive ways.

In the first phase, we compiled the factors into a macro set, showing the
frequency of each one. On examining the entire set, we identified how,
broadly speaking, it comprised three main categories: first, those relating
to the child; second, pedagogy; and third, textual factors, with each of
these including related sub-categories. The naming of the categories is, of
course, a construct on our part, allowing us to represent the groupings of
emphases that the teachers self-identified.

For analysis purposes, we drew on the terms actually used by the

teachers, as recorded on the post-it notes. These have been reproduced
in table form without alteration. The tables indicate the frequency of
references to each factor. Also shown is the accumulative score of each
factor within a category, suggesting its influence relative to other factors
within the same category. Some teachers did not apply a score to every

42 English In Education V01.38 No.1 2004

 Claire Wyatt-Smith and Geraldine Castleton

factor, and therefore such omissions may have diminished the final score
or weighting the index would have received if each factor had been
weighted. Of special interest are the sub-totals of each of the main
categories and what they disclose about how teachers' beliefs about
writing achievement are mediated by implicit assumptions that extend
well beyond academic considerations.

Category 1: 'Child' factors

A distinguishing feature of the 'child' category, shown in Table 1 (p.421, is

how it captures the teachers' wide-ranging knowledge of the child and the
diverse, even hybrid features that teachers reported that they rely on to
come to 'know' the child. There are, for example, influences that pertain
to cultural memberships outside the school, that is, within home and
community, and how factors such as language backgrounds, peer
pressure, and parental influences are perceived to affect the quality of in-
school writing. A limitation of the data is that w e do not have ready access
to information about how the teachers gather information about say, peer
pressure or cultural background. What is clear, however, is that student
achievement in writing is viewed as tied to the students' social situations
outside school, a factor over which teachers have little or no control. Also
clear is how the mix of 'child' factors point to the teachers' interest in
reading and valuing school writing for the insights that it makes available
about the development - psychological and physical - of the writer.

For the teachers, school writing has meaning not only in terms of the
words on the page, but in terms of what it can reveal about how the child is
developing a literate identity, as a student, and more generally, as a social
being. Moreover, the act of writing appears to be highly interactive and
context-dependent. It is where teacher and student meet and where the
teacher makes a reading of the emerging identity of the child, a finding
also addressed by Austin and Freebody (20011 who examined how the
category of 'the child' serves as a criterion of assessment. In our study, the
teachers' stated features serve to indicate how student writing is a site
that exists within a dynamic network of factors. These include, for
example, the child's socio-economic background; the nature and level of
medication the child is prescribed; the perceived level of trust in the
teacher, and whether the child has a high or poor self-esteem -being
confident about taking risks or fearful of making mistakes. Afull listing of
the factors appears in the tables, with the selected maps showing a sub-
set of these.

Interestingly, of all six 'child' factors, the teachers weighted those related
to cultural memberships in home and community as being the most
important in determining writing achievement. This weighting suggests
that teachers were aware of how their work in developing student writers

0 N a t e and Contributors 2004 43

Factors affectina writina achievement

Table 1: Category 1

Factor Frequencv of Accumulated

mention score

CHILD FACTORS AFFECTING WRITING ACHIEVEMENT

AS CULTURAL MEMBER (HOME & COMMUNITY)
Home background: language; social; Family support & encouragement;
environment; importance placed on reading & writing in home; child's
circumstances; parental value placed on learning; Parent's language usage 17 30-31
Everyday experiences of StudenVbackground experience/prior knowledge
(informal); variety of social experiences 12 24-25
Peer pressureiinfiuence; support 7 10-11
Real life writing activity experience; Parental modeling of using reading &
writing in every day life; home early childhood literacy support 4 5
Cultural backgroundfESUlanguage spoken at home 3 8
Parental background and education 1
Parentlcarers' jobs 1
Sub tofal 45 77-80
CHILD AS STUDENT
Student attitude to workldesire to achieve/enthusiasm/motivation/
persistencehot distracted/work ethic 11 29
Interest in subjectltask 5 12
Respect for teacher; relationship with teacher 3 6
Enjoyment of school 2 3-4
Attention to taskhime on task 2 3
Perception of self as learner; confidence in ability to perform 2 2
Ownership of work 1 3
Fear of failure 1 3
Trust in teacher 1 3
Ability to learn from mistakes 1 2
Relevance of task to child 1 2
Behaviour 1 2
Sub total 31 70-71
CHILD AS LITERATE SUBJECT
Skills/prior knowledge (formall: sentence construction; vocab development; 1 34
grammar; purpose/prior knowledge of generic structure; knowledge of 5
punctuation rules; knowledge of language conventions
Love of reading; Read lots of books; independent reading habits; enjoys reading 6 13
Enjoyment of writing; attitude to writing; Values associated with writing 5 13-14
Research skills (dictionary; info seeking.. .) 2 1
Delight in oral language & new vocabulary 1
Early oral language experience 1 3
Child's previous positive experiences with writing 1
Listening to stories 1 2
Sub total 32 66-67
CHILD AS COGNATE BEING
Ability to comprehend instructions; Ascertained/learning difficulties; LO./
intellectual ability; Student abilities/natural ability/ability to think logically 10 25
Reading ability 8 17
Oral language capacity/confidence 4 11
Sub total 22 53
CHILD AS PHYSICAL CONSTRUCT
Fine motor skills 3 6
External context (what happened to the child that day) 2 6
Medication, medical condition (ADHD; Asperges ...I 2 4
Tredness 1 3
Sub total 8 19
CHILD AS PSYCHOLOGICAL CONSTRUCT
Self esteemiconfidence 5 13
Willingness to experiment/make mistakes 1 1
Sub total 6 14

Total 144 299-304

44 English In Education Vo1.38No.1 2004

 Claire Wyatt-Smith a n d Geraldine Castleton

interfaced with social worlds and textual practices outside the school, and
importantly, beyond their control. Moreover, there is the suggestion that
the fit (or lack of fit) between literate practices inside and outside
schooling represents a critical determinant of school writing achievement.
The observation we offer here is that the prospects of success are
optimum where literacy practices in and out of school are 'in tune' -well
aligned one with the other. However, disjunctures are to be expected,
requiring planned review and targeted interventions a t systemic and local
level. Where such explicit provision is not made, there are clear
implications for students' prospects of academic success. This matter is
raised later in the paper, especially in relation to the influence of new
technologies in shaping outside school literacy practices

Category2: 'Pedagogy-centered' factors

Just as the 'child' factors are multi-faceted and serve to locate the identity
of child in the writing, so the 'pedagogy-centred' factors, shown in Table 2
(p.441,focus on the 'personal' attributes of the teacher (enthusiasm and
encouragement), as well as pedagogical style, content, context and
assessment influences. It is in this category that we hear of issues related
to the teachers' roles in modeling and providing explicit assessment
information, as well as student roles in assessment. Teachers identified
factors relating to writing processes and writer interactions, with peers or
with the teacher. Also evident are factors that signal teachers' uptake of
socio-cultural understandings about language in general and their
awareness of key differences between oracy and written language in
particular. There is an emphasis, for example, on the social purposes of
language use, the value of modelling genres in which students are
expected to write, as well as preparation for writing, conferencing
procedures, responding to drafts and explicit teaching of grammar and
vocabulary.

The teachers also put some emphasis on the interactive nature of in-class
assessment practices, with mention made of the nature and extent of
assistance given to student writers, the time allocated to crafting the
writing, and how the teachers monitor performance over time, tracking or
comparing each newly completed piece with previous efforts. Clearly, the
teachers as a group adopted the position that pedagogy and assessment
practices make a difference to writing outcomes, though not nearly as
much, it seems, as cultural memberships, as discussed earlier. To
illustrate this point readers are invited to compare the entries and
weighting of tables 1 and 2.

0 N a t e a n d Contributors 2004 45
Factors affecting wifing achievement

Table 2:Category 2

Factor Frequency Accumulated

of mention score

PEDAGOGY-CENTRED FACTORS AFFECTING WRITING ACHIEVEMENT

Wodeling of genres; Students understand particular genre; Scaffolding and
jeconstruction of genre 5 11
Preparation for writing task by teacherheaching-instruction 4 10
Response to writing stimulus in class/lnvolvement in discussion; Extensive
discussion of ideas before writing 3 6
Students understand functions of grammar; teaching grammar 2 3
Conferencing with drafts 2 5
Tme allowed for vocab. extension/building vocab. Charts 2 6
Teaching punctuation 1 3
Working collaboratively with students to compose 1 3
Practice in oral sentence composition & sharing this with other children 1 3
Sub total 21 50

ASSESSMENT
Content taught in early school under syllabus; opportunities to learn with
previous teachers 4 10
Students understand what is expected; clear expectations and criteria; clear
assessment focus 4 8
Amount of support given to student 3 9
Self-help; has child clarified problems? 1 2
Amount of discussionhesearch time 1 2
How it compares with previous efforts 1 2
Note how long it takes for child to put pencil to paper 1 3
Time given to write text 1 3
Sub total 16 39

TEACHER
Teacher enthusiasm & encouragemenfleacher influence & support/
positive reinforcement 8 15
Teacher competence & knowledge 1 3
Teacher self-review 1 1
Teacher feedback 1 I
Sub total 11 20

TEACHING CONTEXT
IndividuaVgroup (mixed ability; gender grouping.. .) 2 4-5
Classroom atmosphere (noise level.. .); classroom environment conducive to
learning 2 6
Classroom environment:'have a go'
Pressure placed on students to achieve 1 3
Feelings of success for the student 1 3
Sub total 1 2
7 18-19
CONTENT
Embedding genre or topic in meaningful life context 2 4
Purpose of writing 2 6
Expose child to reading of type of genre they are to write 2 6
Audience 1 2
Sub total 7 18
Total 62 145-146

46 English in Education Vo1.38 No.1 2004

 Claire Wyatt-Smith and Geraldine Castleton

Category 3: 'Textual' factors

We name the third main category, 'textual' factors (see table 3 below), this
being a mix of factors: some concerned with surface features of
presentation and amount of writing on the page; some with technical
accuracy or correctness and some with students' attempts at making and
shaping meaning. In terms of accuracy, teachers mentioned, for example,
'adherence to specific generic structure' and 'correct grammar', as well as
other references to structural features at the sentence and paragraph
levels. Also, teachers made references to how good writing is 'good to
read', showing students' 'ability to express themselves through their
written text', and 'evidence of original ideas/creativity'.

The inclusion in this category of 'knowledge of student's previous work' is

of special interest, suggesting how even here, the textual and social
merge. It seems that for the participating teachers, the textual features of
any piece of student writing provided clues about increasing mastery or
control over genres, grammar, punctuation, and vocabulary. It is as
though, as the teachers read each successive piece of writing, they were
able to locate it in its own intertextual history. This suggests how teachers
take an historical approach to reading what individual students write,
lending support to other research showing that as teachers read, they fit
each new piece of writing into a sequence of other texts produced by the
student (Wallace, 1995)and in relation to what Wyatt-Smith (1999) refers
to as the teacher's tacit knowledge files containing a rich set of information
about the student.

Table 3:Category 3
Factor Frequency Accumulated
of mention score
TEXTUAL FACTORS AFFECTING WRITING ACHIEVEMENT
Use of self correcting/editing strategies 7 12
Spelling ability 5 9-1 I
Adherence to specific generic structure 5 14
Punctuation 4 9 11
Presentation 3 5
Evidence of original ideas/creativity 3 8
Understanding of main ideas for paragraphing 2 5-6
Structure of writing 2 5-6
Complex sentence structures 2 2-3
Content of writing appropriate to topic 2 2
Amount of writing on the page 1 2
Correct grammar 1 2
Sequencing of ideas 1
Development of story Iines/ideas 1 3
Verb tense 1 3
Vocabulary quality 1 3
Knowledge of student's previous work/evidence of development 1 1
Use of skills gained in previous units 1 3
Good to read 1 2
Ability to express themselves through their written text 1 2
Total 45 92-98

0 N a t e and Contributors 2004 47

Factors affectinawritina achievement

Overall, the composite of factors in the three categories w e have

discussed to this point indicate that teachers have access to a potent mix
of different ways of knowing (Belenky et al., 1986)writing achievement
and the student writer; ways of knowing that are inherently context
dependent (occurring within the classroom), historical (occurring at points
in time over the school year), and interactional in nature (involving teacher
and student or student and peer/sf. Building on this, w e now turn our
attention to how the teachers themselves represented this knowledge
through their concept maps.

How teachers representedthe factors

We used a computer program, Decision €xp/orer(University of

Strathclyde &Jones, 20001, to represent the teachers' maps. As a tool for
structuring the maps, the program provided a means for managing,
exploring and analyzing the individual thinking that surrounded the
complex issue of writing achievement. Specifically, the program was used
to record the relationship between individual factors, as the teachers
represented them, making it possible for the mapping to record the
reasoning attached to the organization of the factors.

As suggested earlier, we approached concept mapping as an essentially

private act. The understanding was that the exploration of writing
achievement could be contentious, or individuals may be hesitant about
expressing their views in a group setting. Following individual
brainstorming and 'dumping' on paper as discussed earlier, the research
team entered the maps into the Decision Explorerprogram for analysis.
A key feature of the program was that it enabled us to capture those
connections among factors that the teachers identified; a causal link
represented in a single headed arrow, and an arrow head at each end
representing a two-way relationship between factors.

It was not our desire to construct a 'definitive' map, rather to explore

individual maps for evidence of any consistencies or recurring patterns of
beliefs, as well as any inconsistencies. In working with the concept maps,
we were interested in the concepts that the teachers brought forward and
how they represented these in terms of:

C the particular factors that they individually chose to include

C the strength attributed to each factor, gauged as a strong,
moderate or weak link with writing achievement
C the number of directional links back to writing achievement, and
C the number of association links among the factors (association
links are indicated with a two-way arrow).

48 English in Education Vo1.38No.1 2004

 Claire Wyatt-Smith and Geraldine Castleton

Our analysis of all the maps showed that broadly speaking there were
three main types, namely 1) those maps that represented a set of factors
showing onlydirectional links back to writing achievement (that is, each
factor is shown as independent, quite separate from the others in the set);
2) those that represented a set of factors with strong, moderate and weak
linkswith writing achievement, and included some association links within
the set; and 3)those that represented a set of factors with strong,
moderate and weak links with writing achievement, and included
considerable networking of links. For illustrative purposes, a sample of
each of these types of maps has been reproduced, each one being
discussed separately below.

Only directional links

Map One' shows fifteen named concepts, all having a strong link to
writing achievement. In the map the concepts operate independently,
each being a discrete or stand-alone element, though all are tied directly to
achievement. In terms of this style of representation, the map could be
said to be less complex than many others that showed association links
among the concepts, an example discussed later in the paper. The
variation in the selection of concepts included in individual maps and in
how these are depicted is of considerable interest in itself, the starting
position being that there was no intention to value different maps
differently. Instead the diversity of maps was welcomed, suggesting the
mental constructs about writing achievement that teachers have available
to them, but which typically remain unarticulated, existing as latent or in-
the-head knowledge.

So, while Map One is distinctive or unique in the suite of concepts that
it represents and how it represents these, it has the common feature of
showing how, for some of the participating teachers, 1) child-factors
(see references to lunch hour happenings, fatigue, confidence,
managing distractions), 2) textual factors (see reference to structure of
particular genre, tense of verbs), and 3) pedagogical factors (see
reference to the expectations or pressure placed on individuals), were
all taken to impact directly on what students achieve in the writing
classroom. Further, emerging from Map One is what could be
described as a more holistic and situated view of the child-writer who
must draw on a repertoire of inherently linguistic, cultural and personal
resources, even physical attributes. This provides an opening for
considering how, from the teacher's perspective, efforts at improving
learning outcomes may need to adopt a similarly holistic, situated view
that takes account of, but is not restricted to, particular systemic
pedagogical interventions.

'The first table has been deliberately designed to include a side column. The authors'
intention IS to suggest how the tables may have a use in professional development
sessions to generate and record teacher reflections on the maps

0 Nate and Contributors 2004 49

Factors affecting writing achievement

c - c
0 0 0
# # #

50 Engllsh In Education Vo1.38 No.1 2004

 Claire Watt-Smith and Geraldlne Castleton

The teacher's think-aloud transcript of how she actually read and

assessed her own students' writing (see Appendix A, Session 3)
similarly showed how, as the teacher read, she was able to bring to
bear on the writing all three sets of factors. The transcript segment
below shows how the teacher accomplished this, drawing on the
factors as a set of three lenses for reading student writing:

This piece is written by a boy (pause),he's been a reluctant reader

and writer and so I'm very impressed just by the first look at it. The
fact that he's achieved a page of writing in the time and the attempt
to actuallyparagraph now all I have to do is make sure that his, he
knows whyhe'sparagraphingit that way ...

In the segment, the teacher starts from the position of interest in who the
writer is - a boy, a reluctant reader and writer -and then moves to think
about the amount of writing produced and the evidence of the boy's
attempts at paragraphing. The linking of knowledge of the child to talk
about textual factors and then to pedagogicalfactors nowalll have to do
illustrates the teacher's practice of working across all three sets of factors.
This evidence of moving, apparently effortlessly, across the factors
provides an opening for considering the utility of the concept mapping
exercise, as outlined earlier in the paper. Specifically, this exercise could
provide a useful way for teachers to make available, for their own scrutiny,
insights into how they think about writing achievement, and the
implications of this for what they do to secure writing improvement.

Teacher agency and linguistic framing

In carrying forward the critical issue of adopting a holistic, situated
perspective when considering writing achievement and the child-writer,
readers are invited to consider each map presented in the paper not only in
terms of those factors that the teachers identify, but also those factors
over which the teachers could reasonably be said to have control. Of
special interest in this regard is how teachers construe their agency, even
the limits of their authority in securing improved writing outcomes.

Consider how, for example, in Map Two (p.521, priority is given to

pedagogical factors associated with a socio-cultural approach to
teaching and shows a high level of teacher agency. In this map, thirteen
concepts are named, twelve of these characterised as strongly linking to
writing achievement, though once again, no association links are
indicated. The teacher used the term 'genre' in six separate entries,
indicating how effective pedagogy is taken to include explicit teaching of
generic structure and linguistic features (see references to teacher
modeling, scaffolding and deconstruction of genre for example). Also
shown is an emphasis on language use for social purposes (see reference
to embedding of genre in meaningful, life-like context) and how writing

0 N a t e a n d Contributors 2004 51
 Notes
AP TWO
Chld’s abilrty to express
themselves orally dunng and sentencecomplexity used to

aSkmg/answenng questions
I I
\\ 111

write I - II

Teacher modelling the

cessity for lots of d~scussion
talktng through of mung a
Idea5 before wnung
genre Thmkmgaloud

When wnong lndtvlduaUy

ensure use of genenc svucfure
Fmbe&ng the genre in a aod quality of vocab
meanmfl, life lrke context
Scaffolding of genre bemg taught BniIdIng vocab charts and
Lots of deconstructlMlof g e m # of named concepts = 13
= = N m P - = O n
muxedto wnte the gem # of directional links = 13
# of association links = 0

Focal Concept -------- + f--

Teacher-namedConcept 3 strong link 2 (moderate link) 1 (weak link) a link existss association
I I --
 Claire Watt-Smith a n d Geraldine Castleton

performance relates to teacher collaboration with students in writing

processes, and opportunities to engage in reading and speaking activities
(see reference to child's ability to express themselves (sic)orally during
class discussion or when asking/answering questions).

The clear emphasis given in the second map to a socio-cultural approach

could reflect the influence of the English language policy adopted in
Queensland, Australia, and moreover, the influence of syllabus materials
and professional development programs that carried forward the policy
(see Education Queensland, 1994).While the merits of particular policy
directions could be a subject of debate, the map provides clear evidence of
how, at least for some teachers, agency rested with them to secure
quality writing performance by providing effective pedagogy and a
classroom environment conducive to learning. So, while the teacher's
authority as a change-agent is clear, omitted from the map is any mention
of parents or the life circumstances of the child-writer. This feature makes
Map Two strikingly different from the map discussed in the following
section. Before moving on however, it is interesting to note that the
teacher's singular focus on pedagogical factors is also clearly
demonstrated in the 'think aloud' transcript of actually assessing and
judging her students' writing.

As the teacher read and assessed each piece of writing, talking the
process through aloud, she followed an obviously well-rehearsed routine
of focusing first on generic structure, then on features such as
paragraphing, vocabulary, cohesion, grammar, punctuation and spelling.
She also typically commented on whether the writer had been effective in
achieving hidher purpose:

. . . um the secondpiece um the title is 'Our Friend' um again a

recount andsame routine I'm going through and looking at the
generic structure whether it's correct fora recount and in the first
paragraph this child has set the scene. She's got the time and the
place, the characters involved so immediately know where she is
and what she's going to talk about. She then goes on in a time,
again using time sequence. She's got it ordered, urn using good
temporal conjunctions 'about', 'a few days later', 'later aftera
while', so I feel she's got the generic structure of a recount. I'm
now going back to read whether she 's using rich language and use
of again the adverb and the adjective and I feel that she does. I'm
now looking at the grammar andpunctuation and this girl too has
goodsentence structure and most words are spelt correctly. She's
got herpunctuation in place and she 3 finished off with a feeling and
she has moved me as a reader cause again this is one that made me
feel very sad and I think she achieved her purpose. I'd rate that as a
'VHA' (Very High Achievement).

0 Nate and Contributors 2004 53

Factors affectinawritina achievement

Emerging from this talk is how the teacher had ready access to an
internalised set of expectations relating to generic structure, other
linguistic features, and whether the writer has achieved her purpose. Also
clear is how the structural considerations and features that the teacher
names are also not only 'teach-able', but also those that she knew she had
taught, once again confirming her agency in securing writing achievement.
As stated earlier, this was not uniform across the group, and stands in
stark contrast to the concept map and talk presented next.

Different links

Maps Three and Four are more complex representations than those
previously discussed in that they show association links (see two-way
arrows) among the named concepts. The maps suggest the teachers'
sense of how the various concepts do more than simply co-exist. Instead,
they operate in a network, hooking-up one with the other. The variability in
the systems of hook-ups is suggested by the obvious differences
between the two maps, with Map 4 showing ten association links. The
issue here is not the extent to which the teachers' maps could lay claim to
being wholly comprehensive or exhaustive, but rather what they reveal
about the teachers' mind maps of what is entailed in securing writing
improvement.

In Map Three, for example, child factors predominate. There are three
references to background, first in relation to level of background
experiences, second, to home background including breakfast habits and
the timing of going to bed, and third to the stability of the home
background. There are also references to the student's language
background, experiences of literature in the home, and issues relating to
whether the child is medicated appropriately. Against the strength of the
representation of background, the teachers' agency appears to shut down
significantly, with teacher impact restricted to the role of providing quality
resources and to modelling the genre to be learnt.

Evidence of the importance this teacher gave to child factors is

presented in the 'think aloud' around the assessment task conducted in
session two. As the teacher dealt with each of the writing samples
from her class, she clearly identified those children from a non-English
speaking background, often using this as a way of accounting for such
features as the topic selection and mastery over English language,
including structure, tense and spelling:

This one here is another ESL, ah some learning difficulties. He's

also written about sports day urn and he's got less of a grip on
language structure than the ones you've seen so far, urn parts of
it are not even really readable.

54 English in Education Vo1.38 No.1 2004

 riotes
AP THREE
___------____-__-----------
---------______
-- --.
+ - - -
--.
e background. ae
any emoaonal
Degree of teacher modelling
zlS happemng at Performance thathas occurreddunne the
urrent nmei
\\ Achievement j
L I teaching of this genre - I
4
I
I
I Ability to stay on task e.g.
I
If the child is medicated,
bas the child had their
--
Home background e.g. Child's level of
*how late did child go to bed? interest in the g e m Quality of m u r c e s that
*has child had breakfast? have been available for
use when teaching a
# of named concepts = 9
I
particular genre
# of directional links = 9
# of association links = 4
Focal Concept Teacher-named Concept -* -*
I (weak hnk)
-- -- ----*
a link exiytss
f..-----
asociation
*
2 (moderdte link)
strong
Factors affectfnawrltlnna achlevement

In discussing another student's work the teacher commented:

. . . this is quite a, quite a good recount of um what happened in

this camping trip urn but this person's not English as the first
language either.. . and you can tell that from the tenses again
these children have a lot of trouble with tenses.

The 'think aloud' further substantiates the teacher's claims about the
importance of child factors, with a number of the other factors included
in the concept map appearing in the talk. In discussing the writing
sample of another child, the teacher noted, for example:

Right this child is medicated. He is ADD (Attention Deficit

Disorder).. bordering on learning difficulties. He's written about
,

sports day also this is a good effort for him. Yeah six lines but
urn he said what the events were and how he's got past the
markers and challenged other people, what he's won urn lots of
repetitive though 'I', 'I: 'I', 'I' urn so ah I'd urn I'd put 'low
achievement' on that one.

In contrast with Maps One and Two, what is noticeably missing from
Map Three is any mention of the teacher's work in enabling student
writers to increase control of vocabulary, generic structures and other
linguistic features. Instead, writing achievement is represented as being
shaped primarily by the child's cultural membership in the home and
community, as well as by features intrinsic to the person of the child. It
is as though writing achievement is construed as resulting from factors
over which the teacher takes herself to have little, if any influence.

Similarly, Map 4 identifies a suite of only child factors relating to family

situation, behavioural problems, speech language impairments and prior
language learning experiences, all falling outside the present teacher's
direct responsibility. Also of note is how no factor is taken to be wholly
independent or separable from the others. This representation reflects
a conceptualisation of writing achievement as being shaped by a
complex networking of social, cognitive, behavioural and psychological
aspects, with teacher modelling being the only direct reference to
pedagogy as impacting on writing. Such a map once again suggests a
view of the teacher's agency as being limited, even severely, in efforts
to secure real gains in writing achievement. It is as though those other
factors work to situate the teacher's efforts within a complex of social
and cultural realities, with what happens in the writing classroom not
readily separable from them.

56 English in Education Vo1.38 No.1 2004

 Claire Wyatt-Smith and Geraldine Castleton

0 Nate and Contributors 2004 57

Factors affecting writing achievement

The singular focus on child factors shown in Map 4 was also evident in
the transcript of the teacher's think-aloud reading and assessing
session. As the teacher approached the first piece of writing, she
declared:

Okay, first thing I look at is whose piece of writing it is obviously.

I know this student is a good student, so I'm biased already.

The entire transcript showed that it was routine practice for the teacher
to orient to the writing initially through the student's name recorded on
it. It was as though this gave her an anchor for reading, interpreting and
ascribing value to the text. By way of further illustration, consider how,
on moving to the second piece, the teacher commenced the
assessment with:

Okay he's a behaviour problem, actually on Ridilen so he's an

ADD. He's on, he finds it very hard to sit still. Even in this
(referring to test conditions under which the writing was done) he
was up and down all the time and we had to put an aide next to
him just so he would sit still. So I'm amazed that he's actually
written anything. Okay, okay again he hasn't checked his work
through, like he's missed a couple of words, un yeah, again I'm
using previous knowledge of what this kid's like.

Following this initial orienting move on each piece of writing, the

teacher read, actively looking for signs of what she previously knew
about the student, as evident in the following segments:

He invented a game called cricket okay so I'd pick him up on 'call

cricket'. His father is also Italian so with ESL kids or parents you
kind of assume that when they're speaking at home they don't
always use the correct tenses.

The references to cultural and linguistic backgrounds as well as family

matters, health and medication issues, and diagnosed learning
disabilities were recurring features of the transcript. On the one hand,
this makes clear a view of classroom writing as always and inevitably
being impacted by external factors over which the teacher cannot
reasonably be expected to exert control. On the other, we invite
readers to reconsider Map 4 as presenting a somewhat worrying
picture of how, for some teachers, students' background variables
assume such prominence in their thinking and talking about writing
achievement, that they think of themselves as having limited agency in
affecting change.

58 English in Education Vo1.38 No.1 2004

 Claire Watt-Smith and Geraldine Castleton

Before concluding this section, we make mention of four features of

the collective set of thirty-seven maps. While in the published literature,
gender is a variable that is frequently raised as a contributing factor to
writing achievement (Alloway & Gilbert, 1997; Collins, Kenway &
McLeod, 20001, it was identified in only two maps. In one of these,
gender is simply named, while in the other the teacher links gender and
inclination ('boys less inclined'). Though we cannot account for this
omission, it may suggest that teachers do not generally prioritise
gender as a variable affecting writing achievement. This finding is in
stark contrast to the emphasis given in Australia policy and curriculum
documents to gender as powerfully impacting on literacy outcomes.

Also noticeably missing from the collective set of maps were

references to explicitly defined standards and their role in informing
students about expectations of quality. Of the four maps that make
reference to assessment criteria, only one explicitly names standards as
relating to writing achievement. Further, none of the teachers referred
to their own evaluative experience and expertise as a factor, with the
human resource of teacher knowledge about writing quality being
consistently overlooked.

Similarly, none of the maps identified information communication

technologies or more specifically, technological literacy - opportunities
for students to write and read on-line in the classroom. It is as though
accessing and evaluating internet texts, emails, CD Roms and other
multimedia resources - all increasingly common in leisure and work
practices - had not found a place into the teachers' beliefs about factors
influencing in-school writing achievement. There is an urgent need to
redress this situation, responding to the imperative for in-school literacy
practices to have relevance and connectedness to out-of-school
contexts.

Finally, in terms of the push for accountability and measurement data

from Australian governments a t state and national levels, it is of interest
that syllabus documents with a focus on writing were given so little
emphasis while other national and state policy initiatives in literacy
education were totally excluded in the teachers' selections of factors.
The teachers did not mention, for example, current moves to
outcomes-based syllabus documents. Similarly, no mention was made
of moves to measure and report students' writing achievement against
national benchmarks. This suite of omissions could well point to how
the teachers tended to contextualise writing achievement within the
local classroom world rather than in relation to larger, ever changing
global communication and policy contexts.

0 Nate and Contributors 2004 59

Factors affecting writing achievement

Conclusion

This paper has made available a set of factors that teachers have
identified as influences on writing achievement in Year 5. In our
discussion we have drawn on teachers' own concept maps to present
evidence of teachers' direct or first-hand knowledge of the complexity
of writing achievement in schooling and how such complexity arises, as
Murphy (1995) states, 'in part from the nature of the tasks and in part
from the students themselves' (p.257). More specifically, emerging
from the broad categories presented in the first part of the paper and
also from the maps sampled in the second part of the paper we present
a picture of student writing as interpersonal sites within the institution
of schooling where psychosocial, pedagogical and textual factors can
co-exist and play themselves out, having potential to shape whether
students' classroom-based writing is deemed by teachers to be more or
less successful.

The teachers' collective set of factors presented in the paper lends

weight to published writing research, some dating back more than two
decades, indicating how a complex of factors can affect writing quality.
These include: knowledge about the topic; difficulties encountered
during the composing process; and attitudes to writing covering general
self-esteem, attitudes to particular topics, and one's confidence as a
writer (see for example, Baker & Quellmalz, 1981; Freedman, 1983;
Graves, 1978; Shaughnessy, 1977; Wyatt-Smith, 2000). The
contribution of our discussion is to challenge teachers, teacher
researchers and those responsible for education policy to make an
opening for teachers' ways of knowing achievement, attending to how
they intersect with formal or authorized ways of knowing, as captured
in official education guidelines and policy. This is not to call into
question the legitimacy of such teacher insights, or to fuel the
argument that teachers' notions of achievement are arbitrary, being
subject to bias or whimsy. It is instead to put forward the argument
that the absence of explicit attention to such ways of knowing is
potentially self-limiting for the education community, not to mention the
obvious impact of how well existing systems serve students of varying
abilities.

Given the finding that curriculum and assessment policy exerted a fairly
limited influence over the sets of factors that the Australian teachers
identified as affecting student writing achievement, it is clear that much
more also needs to be learned about how policy impacts on teacher
beliefs and practices, as they relate to pedagogy and assessment.
There can be little confidence in claims about accountability and
transparency of assessment practices until teachers' latent or in-the-
head knowledge and beliefs about writing, the writing classroom,

60 English in Education Vo1.38 No.1 2004

 Claire Wvatt-Smith a n d Geraldine Castleton

themselves as teachers, and the role of the student writer, are regarded
as potent factors in how teachers view writing achievement.

Finally, w e argue for the need to work constructively with teachers as

they bring to the surface their beliefs about writing achievement,
shown here to operate within a complex of implicit and in some cases,
wide-ranging, assumptions. In this discussion w e have shown how
these beliefs extend to the attributes of the child, to pedagogy and to
textual considerations, with all of these being capable of linking, one
with the other, in a myriad of combinations. If w e accept that such
beliefs have the potential to shape teacher-student interactions, and
beyond this, to inform how teachers seek to improve students' writing
outcomes, then there is much to be gained by enabling teachers to
open for scrutiny the nature of those beliefs. In giving more overt
acknowledgement of teachers' ways of knowing, there is the promise
of establishing more appropriate and rich indicators of achievement and
for addressing the issue about the real agency of the teacher in
improving writing outcomes.

Acknowledgement: Thanks are due to Professor Ray Cooksey,

University of Armidale, for his work in developing the maps from the
teachers' paper copies.

References
Alloway, N. & Gilbert, P. (1997) Boys and literacy: Lessons from
Australia, Gender and Education, 9 (11, 49-58
Austin, H. & Freebody, P. (20011 Assembling and assessing the 'child-
student': 'The child' as a criterion of assessment, Journal of Curriculum
Studies, 33 (51, 535-550
University of Strathclyde & Jones, M. (2000) The Decision Explorer
Program, Glasgow, Scotland: Banxia Software Ltd.
Baker, E., & Quellmalz, E. (1981) Effects of visual or written topic
information on essay quality, in E. Quellmalz, Test design: Studies in
writing assessment, Los Angeles: Center for the Study of Evaluation,
University of California Los Angeles
Belenky, M.F., Clinchy, B.M., Goldberger, N.R., & Tarule, J.M. (1986)
Women's Ways of Knowing: The development of self, voice, and mind,
New York: Basic Books
Castleton, G., Wyatt-Smith, C., Cooksey, R. & Freebody, P. (2003) The
nature of teacher's qualitative judgements: A matter of context and
salience, Part two: 'out-of-context' judgements, Australian Journal of
Language and Literacy, 26(2), 33-42
Collins, C., Kenway, J. & McLeod, J. (2000) Factors influencing the
educational performance of males and females in schools and their
initial destinations after leaving school, Deakin University, Geelong &
University of South Australia, Adelaide

0 N a t e a n d Contributors 2004 61
Factors affecting wrltlng achlevement

Freedman, S.W (1 983)Student characteristics and essay test writing

performance, Research in the Teaching of English, 17(4),31 3-325
Graves, D. (1 978) Balancing the basics: Let them write, New York: Ford
Foundation
Halford, G. (1 993)Children's Understanding: The Development of
Mental Models, Hillsdale, NJ: Lawrence Erlbaum
Katz, S. (2000)Competency, epistemology and pedagogy: curriculum's
holy trinity, The Curriculum Journal, 1 1 (2),133-I
44
Kinchin, I. & Hay, D (2000)How a qualitative approach to concept map
analysis can be used to aid learning by illustrating, Educational
Research, 42(1),43-58
Kress, G. (1 985)Linguistic processes in sociocultural practice, Deakin,
Victoria: Deakin University Press
Murphy, P. (1995) Sources of inequity: Understanding students'
responses to assessment, Assessment in Education: Principles, Policy
& Practice, 2(3),249-270
Novak, J. (1990)Concept mapping: a useful tool for science education,
Journal of Research in Science Teaching, 27(1),937-49
Novak, J. (1 998)Learning, Creating and Using Knowledge: Concept
Maps as Facilitative Tools in Schools and Corporations, Hillsdale NJ:
Lawrence Erlbaum
Sadler, D.R. (1989)Formative assessment and the design of
instructional systems, lnstructional Science,18,119-144
Shaughnessy, M. (1977)Errors and expectations: A guide for the
teacher of basic writing, New York: Oxford University Press
Wallace, C. (1 995)Reading, Oxford: Oxford University Press
Wyatt-Smith, C. (1 999)Reading for assessment: how teachers ascribe
meaning and value to student writing, Assessment in €ducation, 6 (2),
195-223
Wyatt-Smith, C.M. (2000)Exploring the relationship between large-scale
literacy testing programs and classroom-based assessment: a focus on
teachers' accounts, Australian Journal of Language and Literacy, 23
(21,109-1
27
Wyatt-Smith, C., Castleton, G., Freebody, P. & Cooksey, C. (2003) The
nature of teacher's qualitative judgements: a matter of context and
salience Part one: 'in-context' judgements, Australian Journal of
Language and Literacy, 26 (2),11-32

62 English in Education Vo1.38No.1 2004

 Claire Wvatt-Smith and Geraldine Castleton

Appendix A:

Session 2: Each teacher is paired with a research assistant to tape record

'a think-aloud' session. In the session, the teachers are encouraged to
speak out the factors they are thinking about as they mark each of the
student samples they have brought from their own classrooms. The final
mark allocated is recorded and photocopies of these writing samples are
later taken for the project records. It is stressed that the purpose of the
think-alouds is to record the factors the teachers are actually taking into
consideration in the marking process. The teachers' own class samples
are referred to as 'in context' samples, as they know the students and the
pedagogy involved in the production of the writing samples.

Session 3: In this session the teachers look at the 25 samples chosen to

reflect a variety of abilities and types of writing collected from a variety of
year 5 state school classes. The samples had been collected in Phase One
of the study'. The teachers are asked to use the same grading system
they used on their own students' scripts to assess these 'out of context'
samples. A limited amount of background information on the way in which
these pieces were produced ...e.g. 'draft copy; some class time research;
historical diary', is provided. As with the previous session, each teacher's
think-alouds are recorded individually.

Session 4: In this session the teachers revisit the 25 'in context' and 25
'out of context' student samples, assessing them on this occasion against
the national benchmark for writing, Professional Elaboration, Year 5.

Session 5: A brief revisit of the cause mapping occurs in this session to

allow the teachers to confirm or amend their initial selection of writing
performance factors. Also, this is the moderation session in which
teachers are able to hear and discuss the judgments of the other teachers
on one sample of their students' work. To the other teachers, this is an
'out of context' judgment (unless the teacher has a teaching partner
present). The conversations reconciling the 'out of context' with the 'in
context' judgments are tape-recorded.

' The Chief Investigators of the study are: Professor Peter Freebody and Associate Professor
Claire Wyatt-Smith, both of Griffith University's Centre for Applied Language, Literacy and
Communication Studies, and Professor Ray Cooksey, University of New England.

0 Nate and Contributors 2004 63