ArticlePDF Available

Abstract and Figures

Increasing attention has been paid in recent years to the ways writing may engage adolescents in higher levels of epistemic complexity (i.e., postulating causes, reasons and other relations or theories related to scientific phenomena), yet in secondary science classrooms, writing has primarily been used for assessing students' content knowledge. Embedded in a larger national study of secondary writing in the United States, this study investigated the qualities of science writing samples collected from 33 adolescents attending schools identified for exemplary writing performance. We asked: How is epistemic complexity reflected in adolescents' writing?; How does the level of epistemic complexity differ by adolescents' language background, grade level, and school context?; What is the nature of the relationship of types of writing and higher or lower levels of epistemic complexity? We found the majority of writing adolescents produced did not show evidence of high levels of epistemic complexity. Notable exceptions were reading reflections and lab reports. Implications for adolescent science writing instruction are discussed in light of higher standards for disciplinary writing in secondary schools.
Content may be subject to copyright.
Epistemic Complexity in Adolescent
Science Writing
Kristen Campbell Wilcox*, Fang Yu*, & Marc Nachowitz**
*University at Albany | USA, **Miami University of Ohio | USA
Abstract:
Increasing attention has been paid in recent years to the ways writing may engage
adolescents in higher levels of epistemic complexity (i.e.,
postulating causes, reasons and other
relations or theories related to scientific phenomena), yet in secondary sci
ence classrooms, writing
has primarily been used for assessing students’ content knowledge. Embedded in a larger national
study of secondary writing in the United States, this study investigated the qualit
ies of science
writing samples collected from 33 ad
olescents attending schools identified for exemplary writing
performance. We asked: How is epistemic complexity reflected in adolescents’ writing?; How
does the level of epistemic complexity differ by adolescents’ language background, grade level,
and school context?; What is the nature of the relationship of types of writing and higher or lower
levels of epistemic complexity? We found the majority of writing adolescents produced did not
show evidence of high levels of epistemic complexity. Notable exceptio
ns were reading reflections
and lab reports. Implications for adolescent science writing instruction are discussed in light of
higher standards for disciplinary writing in secondary schools.
Keywords: Adolescent writing; science writing; epistemic complexity; English learners
Wilcox, K. C., Yu, F., & Nachowitz, M. (2015). Epistemic Complexity in Adolescent Science
Writing. Journal of Writing Research, volume(issue), ##-##.
Contact: Kristen Campbell Wilcox, University at Albany/School of Education, Albany NY 12025, |
USA; email: Kwilcox1@albany.edu.
Copyright: Earli | This article is published under Creative Commons Attribution-Noncommercial-
No Derivative Works 3.0 Unported license.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 2
A growing body of literature highlights the importance of writing in the development of
21st-century dispositions and skills that involve reorganizing and generating new
knowledge (Bereiter & Scardamalia, 1987; Chuy, Scardamalia, & Bereiter, 2012; Langer
& Applebee, 1987). However, while a number of scholars have provided a convincing
case for how writing competence is fundamental to fostering these 21st Century
dispositions and skills (Hand, Lawrence, & Yore, 2010; MacArthur, Graham, &
Fitzgerald, 2008; Moje, 2011; Norris & Phillips, 2003; Rivard, 1994), as several recent
studies have shown, secondary level science teachers typically pay little attention to
writing or to the potential different kinds of writing tasks might have on students’
development of both writing competencies and content knowledge (Rijlaarsdam,
Couzijn, Janssen, Braaksma, & Kieft, 2006; Wellington & Osborne, 2001).
Furthermore, as Pearson et al. found (2010) many secondary science teachers see
reading and writing as universal skills that are developed elsewhere (namely in English
Language Arts classrooms) and do not understand how to use writing to teach their
adolescent students the unique ways in which meaning is communicated in the
scientific community.
This scenario is changing however. In the United States, the Common Core State
Standards (CCSS) for literacy (adopted by the majority of states) emphasize writing in
the core disciplines including science (National Governors Association Center for
Best Practices & Council of Chief State School Officers, 2010). Specifically the CCSS
stress developing students’ abilities to examine and convey complex ideas clearly and
accurately; produce writing appropriate to different purposes and audiences; and draw
evidence from sources to support claims (National Governors Association Center for
Best Practices & Council of Chief State School Officers, 2010). In addition, the recently
published Next Generation Science Standards (NGSS) require students to engage in the
practice of “obtaining, evaluating and communicating information” (Achieve, 2012).
Both the CCSS and the NGSS mark a shift in the emphasis being placed on disciplinary
writing and highlight what some studies on writing in secondary science classrooms
have suggested: Writing is an important, perhaps critical, component of learning to
“do” science and think like a scientist (Hand & Prain, 2002; Metz, 2006; Porter et al.,
2010).
In science, providing a claim and evidence of a claim, for example, represents a
particular way of knowing (i.e., epistemology). However, to what extent science
teachers and their students understand the relationships of what they write and how
they write to the work of scientists in reorganizing and generating knowledge has come
under question. Scholars such as Prain and Hand (1999) have noted students oftentimes
demonstrate a “limited capacity to explain how knowledge claims are established in
science in relation to learning through writing, or to understand how writing could act
as an epistemological tool” (p. 160). These findings suggest that students need more
support in engaging in writing that can function as an instrument for knowledge
reorganization and generation as in and through these writing activities the
3 | JOURNAL OF WRITING RESEARCH
epistemologies undergirding scientific disciplines become enacted. The concept of
epistemic complexity becomes salient here.
Since we were interested in the qualities of writing adolescent students were
producing in their science classes with a particular concern for ways scientific
knowledge was being represented in their written work, we turned to the concept of
epistemic complexity following the work of Hakkarainen (1998, 2003) and Zhang et al.
(2007). In this vein of inquiry, epistemic complexity provides a way to characterize
how writing functions as an instrument for knowledge representation. Epistemic
complexity can be defined as a tool to measure the extent to which a writer explains
phenomena, postulates causes, reasons and other relations or theories related to
scientific phenomena (Hakkaranien, 2003; Kelly & Takao, 2002; Kuhn, 1993; Salmon,
1984). Epistemic complexity provides a way to index science writing on a scale from
simple descriptions of scientific phenomena to increasingly more complex explanations
of relationships of phenomena and analyses of phenomena that include arguments with
claims and counterclaims, supporting evidence and warrants.
In sum, although the CCSS and NGSS raise expectations for writing in American
secondary science classrooms, what kinds of writing adolescents should be expected to
do in order to align to these standards is less clear. In an effort to explore this concern
empirically, the study reported here sought to (1) characterize the extent to which
adolescents’ writing exhibited high levels of epistemic complexity (i.e., writing
involving postulating causes, reasons and other relations or theories related to scientific
phenomena), (2) determine whether the writing of students of different language
backgrounds, grade levels, and school contexts differed with regard to levels of
epistemic complexity evident in their writing, and (3) characterize the types of writing
that exhibited higher levels of epistemic complexity. We pursued these questions in
order to inform future research and efforts to improve the teaching of science writing in
secondary school classrooms.
1. Related Literature Science Discourse: Its Characteristics and
Epistemological Foundations
The goal of science education may be considered twofold: not only mastery of
scientific content and concepts but also learning how to engage in scientific discourse
as to represent and generate new knowledge as expert scientists do (Bricker & Bell,
2009). In this section we examine the literature establishing how knowledge
representation and generation is viewed in the scientific domain.
Descriptions of observable phenomena and explanations of theory are common
elements of scientific discourse (Kuhn, 2010, Rijlaarsdam et. al., 2006). However,
teaching students to describe phenomena and explain theory without knowledge of
how they are related to each other is problematic for a variety of reasons. As Kuhn et
al. (2008) explain, contemporary science education emphasizes the importance of
developing scientific reasoning skills as well. Theoretically, through the development of
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 4
reasoning skills, students would learn to clarify and reorganize their understandings of
scientific phenomena and theories and thus enhance their understandings of them.
While doing this work, they will also deepen their understandings of the
epistemological foundations of science.
Learning to engage in discourse that involves scientific reasoning requires learners
to not only describe phenomena or explain a theory, but also connect pieces of
information and describe causal relationships (Hempel, 1965; Salmon, 1978). To do so,
as Salmon (1978) asserts, advanced science discourse is over and above mere
description and explanation as it embraces two powerful processes. First, explanation
of a phenomenon “…essentially involves locating and identifying its cause or causes”
(p. 685). Causal relationships are typically established directly from observation in
various contexts rather than a single one and in turn these observations are likely to
generate the second processsubsumption under law” (p. 685). On a higher cognitive
plane, general law arises from theoretical science which not only explains a
concrete/observable phenomenon or theory, but also predicts what may happen under
a specific situation and the possibility of some outcome or alternatives. As scientists
articulate these cause-effect sequences and their potential outcomes, they engage in
reorganizing and generating new knowledge. For science learners, engaging in writing
that requires articulations that go beyond mere descriptions of phenomena or theories
can afford them the opportunity to comprehend and elaborate the hidden theoretical
mechanisms of the material world, and thus, reorganize and generate science
knowledge.
In a synthesis of research examining epistemological understanding in science,
Kuhn et al (2000) posit that there are four levels (i.e., realist, absolutist, multiplist, and
evaluativist) moving from subjective understanding in the immature learner through a
balance of objective and subjective understanding in the more mature learner. A realist
perspective does not involve critical thinking, yet the evaluativist is characterized by
seeing knowledge as something generated by human minds and that which may be
uncertain, requiring judgments that promote and even require sound assertions.
Applying these levels to a study of learners ranging from elementary school through
adult, the authors conclude that “reasoned argument is worthwhile and the most
productive path to knowledge” (p.325).
While some scholars note that engaging younger learners in higher-order critical
reasoning is difficult and oftentimes not addressed explicitly in coursework (Duschl,
2008; Sandoval, 2005), adolescents can learn the complex cognitive processes of
scientific discourse by explicit instruction in claim and counterclaimboth supported
by evidence through argument and dialogue (Kuhn et al, 2009). Building on earlier
research establishing that instructional emphasis on explanation may conflict with
students’ attention to the importance of evidence in justifying claims and observations
(see, for example, Brem & Rips, 2000; Kuhn 1993), Kuhn and Crowell (2011) tested an
intervention in which students were taught to engage in argumentation, explicitly being
instructed in the skills and importance of claims and counterclaims justified with
5 | JOURNAL OF WRITING RESEARCH
evidence. Findings suggested that students who engaged in dialogic argumentation
demonstrated a higher quality of scientific reasoning in addition to greater awareness of
the relevance of evidence to argument in scientific discourse. The authors conclude
that the dialogic method of argumentation is valid for developing the cognitive skills
required for scientific reasoning.
Research on the teaching of argument as a central component to how knowledge is
constructed in the scientific community has been of particular interest to some scholars
in the past couple of decades (Bicker &Brell, 2009). For example, the Toulman
Argument Pattern (TAP) has been investigated with regard to how it might assist
students in understanding scientific constructs and for assessing the quality of their
written argumentation (Osborne et al., 2004). Building on Toulman’s emphasis on
warrants and claims in science writing, Konstantinidou and Macagno (2013) theorized
that argumentation schemes can be used for helping students analyze, reconstruct, and
improve their reasoning skills, particularly as they adjust new understanding to connect
with prior knowledge on a specific issue. The authors note that argumentation schemes
are particularly applicable to science education in a two-step process. First, an
argument is analyzed with the claims supporting the conclusion, and, second,
constructing an argument requires that links to evidence and prior knowledge are
identified, retrieved, and defined. In this multi-step process, “the student is requested to
analyze and reflect on the notions underlying his reasoning about a specific scientific
phenomenon” (p.1085).
This body of scholarship highlights that both explanation and argumentation are
essential elements of how scientific knowledge is expressed, and that teaching students
to engage in writing that requires these actions also involves engaging in the kinds of
complex reasoning activities that more expert scientists do. As the studies highlighted
here suggest, some kinds of writing activities have the potential to help students
understand both what constitutes scientific understanding as well as develop the
dispositions and skills to engage in advanced scientific discourse grounded in deep
epistemological understandings.
1.1 The Nature of Science Writing Taught in Secondary School
How well do the types of writing adolescents do advance these deeper epistemological
understandings and align with developing 21st Century dispositions and skills in science
as described in the CCSS and the NGSS? The relevant literature in this area is
dominated by a focus on one particular type of writing: The laboratory report. The lab
report has been of concern to researchers as it is a common type of writing required of
secondary- and post-secondary level students and essentially provides an outline for a
particular technique, approach, and reasoning process called the scientific method. In
one study of eighth grade students’ experiences with writing lab reports, Keys (1998)
revealed that although students generated hypotheses, examined patterns in data, and
made general knowledge claims in response to the task of a lab report as they were
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 6
explicitly instructed to do, the fixed structure of the task also constrained students’ deep
thinking into scientific problems. The aforementioned finding is problematic since as
discussed in a report commissioned by the National Academy of the Sciences and the
National Science Foundation, one of the purposes of the laboratory experiment is to
promote an understanding of the complexity and ambiguity of scientific knowledge or
what aligns to the earlier-mentioned “evaluativist” perspective (Singer, Hilton, &
Schweingruber, 2006).
With an interest in the potential of offering scaffolded writing tasks in prompting
students to engage in more complex thinking in their writing, Hand, Wallace, and
Yang (2004https://mail.google.com/mail/?shva=1 - 13df05bd404eb242__ENREF_4)
sought to identify the outcomes of infusing 7th grade science laboratory instruction with
science writing heuristics (prompting students to ask such questions as “what are my
questions?”, “what did I do?”, “what did I see?”,“what can I claim?”). The results of their
analysis of students’ retrospective accounts of writing in response to these questions as
part of their laboratory report task suggest that such an approach positively impacted
students’ understandings of the rhetorical features of a scientific claim and argument
and that such writing enhanced their learning of the science content. This research
draws attention to the potential for such heuristics to encourage better writing and
deeper thinking about scientific content in contrast to the less promising standard lab
report assignment.
The body of research we have explored here illustrates that writing in science can
foster both higher levels of scientific reasoning and understanding of content. However,
very little is known about the qualities of secondary school students’ science writing in
a variety of school contexts and among students in different grade levels and from
different language backgrounds. Further, no studies have investigated the writing of
adolescents in schools with histories of exemplary writing performance as to provide
potential exemplars of adolescent writing that align with the higher standards for
disciplinary writing as described in the CCSS and NGSS.
1.2 Theoretical Framework
Sociocultural theory provides a lens through which we may deepen our understanding
of scientific writing and the contexts that produce varying levels of epistemic
complexity in student writing. A growing body of literature has focused on investigating
science education from a sociocultural perspective (Green & Dixon, 1993; Jimenez-
Aleixandre et al., 2000; Kelly & Bazerman, 2003; Kelly, Chen, & Crawford, 1998; Kelly
& Crawford, 1997; Lemke, 1990). By adopting methods including ethnography,
discourse analysis, ethnomethodology, and others these studies understand that
learning science is a sociocultural activity where disciplinary knowledge is constructed
in a community culture through a variety of oral, aural, visual, and written activities.
This perspective views members of the scientific community as ascribing meaning to
the processes, artifacts, practices, and signs and symbols that they construct in and
7 | JOURNAL OF WRITING RESEARCH
through their activities and in discourse traditions that have developed over time and
oftentimes in unique ways.
Discourse plays an important role in socializing learners into a disciplinary
community. As Lemke (1990) explained, speaking, writing, drawing, calculating, and
experimenting, are conduits through which the "conceptual systems" and the "scientific
theories" are taught and learned. Thus, science writing (along with other forms of
discourse), from a sociocultural perspective, is seen as a potentially powerful activity
for developing understandings of the epistemological foundations of science.
Conceiving of science as a "special way of talking about some set of topics" (Lemke,
1990, p. 155) connects the learning of science to learning the particular uses of
scientific language. Mastering scientific discourse in ways that allow for generation of
new knowledge involves more than describing content, concepts, or theories, as Lemke
explained, "it is a matter of the ways these special words are used together, the
semantic relations we construct among them when we use them" (p. 155).
In alignment with a sociocultural framing that takes into account the ways
discourse reflects and is embedded in the development of complex ways of
understanding science and generating new scientific knowledge, we investigated the
following research questions: (1) How is epistemic complexity reflected in adolescents’
writing?; (2) How does the level of epistemic complexity differ by adolescents’
language background, grade level, and educational context?; and (3) What is the nature
of the relationship of types of writing and higher or lower levels of epistemic
complexity?
2. Method
The current study was embedded in a national study of writing instruction which
researched the teaching and learning of writing in middle and high school settings
across the United States. The national study aimed, in large part, to track changes in
approaches to writing instruction from a study conducted in 1981 (Applebee), and to
investigate the extent to which student characteristics (e.g., achievement histories,
grade level, language background) and school contexts relate to different writing
experiences and outcomes.
2.1 The Larger Study Sample
Criteria used to identify the sample for the larger study including (1) diversity in state
requirements for writing and (2) schools’ histories of performance in writing. With
regard to diversity in state contexts, the larger study sought to include states that had
different requirements for writing in their high stakes exit-level assessments: California,
Kentucky, Michigan, New York, and Texas were chosen for this reason. Within this
sample, New York was the only state that required writing of a paragraph or more in
science and mathematics on the secondary exit exam. Kentucky offered the option to
include science writing as part of a portfolio assessment and other states included
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 8
writing in English language arts only and each with slightly different emphases in genres
expected. Second, since one of the objectives of the larger study was to investigate
potentially better-case scenarios for secondary level writing, schools with histories of
exemplary writing performance on exit-level exams that were also nominated by
leaders in the field of English as having traditions of excellence in the teacher of writing
were considered for inclusion. Unlike traditional corpus studies that seek to attain a
representative sample of writing, the intent of this sampling method was to highlight
writing produced in a variety of diverse, yet unique schools in terms of being
characterized as historically exemplary in English language arts performance.
2.2 Current Study Sample
Because one of our interests was in the experiences of both native English speaking and
English learner students, of the five states included in the larger study, this study drew
on data collected in the three most linguistically diverse states: New York, California,
and Texas. Both Michigan and Kentucky were eliminated from the current study since
the sample of students from those states did not include both native English speaking
and English learner participants in all of the grade levels of interest (6th, 8th, 10th, and
12th). Therefore, while the larger study included 14 students from California, 20 from
New York, and 36 from Texas, we selected 11 students from each of the three states,
totaling 33 students on which to focus our analyses. This purposive sample took into
account the following criteria: The total number of pieces of writing from students in
the current study was representative of the total number of pieces of writing produced
by students in the larger sample, inclusion of all target grade levels (i.e., 6th, 8th, 10th,
and 12th), both native English speakers (NES) and English learners (ELs), and both males
and females (see Table 1).
The schools these students attended varied in size and demography as can be seen
in Appendix A. Of these schools, several had support from school administration in
implementing writing across the curriculum programs (e.g., Albert Leonard in New
York) and others had faculty with strong ties to the National Writing Project which
provided ongoing support for the development of teachers’ writing pedagogy (e.g., King
Drew in California). Two schools, stood out from the others in the extent of their ties
with external, teacher professional development: King Drew High School in California,
a magnet school with a science and health theme had particularly close and ongoing
ties with the University of California Los Angeles (UCLA) Writing Project. Grisham
Middle School, like all the schools in the Austin Unified School District, had extensive
support from the local Math/Science Collaborative run by the local university and
featured extensive and ongoing faculty training in teaching laboratory report writing
from grades one through twelve.
9 | JOURNAL OF WRITING RESEARCH
Table 1. Participant Characteristics
School
State
Grade
Pseudonym
Language
Background
Tota l
Montebello
CA
6
Alissa
EL
8
Angel
EL
8
Emily
EL
6
Lisa
NES
John Adams
CA
8
John
NES
King Drew
CA
12
Bob Bill
NES
10
Kobe
NES
12
Paris
NES
10
Sunny
EL
12
Guitar player
EL
10
Arial
NES
Grisham
TX
8
G6
EL
6
G7
NES
8
G2
EL
6
G8
NES
Spring Branch
TX
8
SB8
NES
6
SB5
EL
8
SB12
NES
McCallum
TX
10
M3
NES
10
M1
EL
Round Rock
TX
12
RR6
EL
12
RR1
NES
Albert Leonard
NY
8
Betty
NES
6
LouAnn
NES
Port Chester
NY
8
Ya s m i n e
EL
6
Tony
EL
8
Karen
NES
Batavia
NY
12
Don
NES
10
Randy
NES
10
Dave
NES
10
Chin
EL
New Paltz
NY
10
Shane
EL
12
Shanice
EL
2.3 Data Collection
The 33 students in this study produced 304 pieces of writing in their science classes
over an approximately 13 week term (half of a school year). This writing included such
writing as worksheets, short- answer responses, and class notes as well as more
extended writing such as lab reports. Collection procedures were adapted to the
particular relationships at each school site. In some cases the on-site coordinator
collected the work from subject-area teachers on a regular schedule; in others, the focal
students brought their work individually to the on-site coordinator for forwarding to the
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 10
research team. In either case, the originals were returned to the students and copies
forwarded on a regular basis to the field researchers. As data were received at the
research center, they were inventoried by staff and entered into SPSS as described
below and in further detail in the larger study methods and procedures document
available online (Author, nd.).
2.4 Data Analysis
Our procedure for the analysis of students’ written work was four-fold. To begin, the
students’ writing was first categorized by type based upon Applebee’s 1981 study
which framed the larger national study of which this one is part. Types were defined as
mechanical, informational, personal, and imaginative (see Appendix B). Next, based on
hierarchies for epistemic complexity used in previous studies (e.g., Hakkarainen, 1998;
Webb, 2002), one of the research team members coded the 304 pieces writing on a
five-point scale with writing at the lowest level (i.e., level one) showing evidence of
separated pieces of information and writing at the highest level showing evidence of
postulating causes, reasons and other relations or theories related to scientific content.
The two highest levels (four and five) are where explanations of phenomena that might
include arguments with claims and counterclaims, supporting evidence and warrants
are evidenced. The categories of epistemic complexity we used in this analysis, which
are slightly modified from those used by other researchers, are defined in Table 2.
Table 2. Levels of Epistemic Complexity
Level
Definition
1
Separated pieces of facts. A statement consisting of a list or table of facts with hardly any
integration or connections.
2
Partially-organized facts. A statement consisting of facts that were loosely organized
together. The facts were stated without relating them to each other by means of causal or
some other connections. Only a minimal amount of inference seemed to be involved.
3
Well-organized facts. A statement consisting of rather well-organized factual or
descriptive information. Although the ideas did not explicitly provide an explanation, it
was meaningfully organized and had a potential of facilitating understanding of the issue
in question.
4
Partial explanation. A statement represents an explicit attempt to construct an explanation
and to provide new information, but the explanation was only partially articulated. It was
only an explanatory sketch that was not further elaborated.
5
Well organized explanation. A statement containing postulations of common causes,
reasons and other explanatory relations, or theoretical entities.
Of all of the written work, 17.4 % could not be categorized on the epistemic scale
because these pieces were not legible or they were selected-response items such as
11 | JOURNAL OF WRITING RESEARCH
multiple choice or matching exercises that did not require students to produce text that
could be analyzed using the scale of epistemic complexity.
In the next stage of analysis, we used a concept-mapping procedure that was
originally constructed by Chi et al. (1981), and was employed to represent semantic
node-link networks of key terms (see Appendix C). Proceeding from Chi et al., Fellows
(1994) adapted a concept-mapping approach to transform students’ writing into
representations of their structure through tracking essential ideas and relationships
among these ideas. Similar to Chi et al. and Fellows (1994), we employed concept-
mapping to transform students’ writing into visual representations of their structure.
Two types of nodes were identified in this process, and they are: a) entity (i.e., the key
concept(s) in a statement- indicated by circles); and b) relationship (i.e., the association
between/among different entities- indicated by diamonds). As can be seen in Figure 1,
the following excerpt from a student’s work was a level 2 characterized by two or
multiple entities that had only one layer of relationship. For example, the following
excerpt was indexed at Level 2: Partially-organized facts:
Dermal bones form in subcutaneous membranes. They’re mostly composed of
cancellous[sic], bone and parts of irregular bones. (Student 1)
Figure 1. Example of level 2: Partially organized facts
To establish reliability of the coding procedures used in the study, two university
teachers in the Department of Education independently coded 20% of the written
samples (randomly selected) (Zhang, Scardamalia, Reeve, Messina, 2009). We used
Cohen’s Kappa to calculate the inter-rater agreement for each pair of raters, and then
took the average of Kappa across all pairs of raters (Griffin, 2011), resulting in an inter-
rater reliability of 0.89 (SE = 0.02).
To further examine the patterns of epistemic complexity and answer our second
research question, we used a One-Way ANOVA to test the statistical significance with
regard to students’ language background, grade level, and school context
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 12
acknowledging that the statistical power of such a test on such a small sample is weak.
The unit of analysis is participant (n=33), whose level of epistemic complexity (EC) was
defined as dependent variable, and language background (LAG), grade level (grade),
and school context (STA) were independent variables. Each participant’s level of
epistemic complexity (EC) was calculated as a weighted average, as the number was
the number of written items in a level, and the weight was the level of epistemic
complexity (1-5). For example, if a student produced 6 items at level 1, 5 items at level
2, 20 items at level 3, 1 item at level 4, and 2 items at level 5, then his weighted
average of the level of epistemic complexity is:
(6*1+5*2+20*3+1*4+2*5)/(6+5+20+1+2)=2.65.
In our sample, five students had fewer than 2 pieces of writing, therefore, we ran
the ANOVA tests again to exclude the 5 students and compared the results with the
larger sample. We also used SPSS to examine whether there was a normal distribution
across the data set, and results showed that the data met the assumptions of ANOVA
(data were normally distributed).
Finally, once we identified patterns of epistemic complexity across the writing of
students with different language backgrounds, grades, and school contexts, we crafted
descriptive illustrative cases of epistemic complexity (Yin, 2005).
3. Findings
In response to our first research question (How is epistemic complexity reflected in
adolescents’ writing?), we found that little of the writing in our sample from schools
with histories of exemplary performance in writing represented higher levels of
epistemic complexity. Overall the average level of epistemic complexity in students
writing was 1.7 (between separated pieces of facts and partially organized facts) on the
scale (Level 1: Separated pieces of facts; Level 2: Partially-organized facts; Level 3:
Well-organized facts’ Level 4: Partial explanation; Level 5: Well organized
explanation).
We used the categories of mechanical, informational, personal, and imaginative
(defined in Appendix B) to parse the sample. Table 3 shows the break-down of levels of
epistemic complexity by these categories.
Within the category of mechanical writing, more than half of the sample fell into
the first level of epistemic complexity in the subcategories of short answer questions,
fill-in-the-blank exercises, and symbolic representation. Short answer questions were
the only type of mechanical writing at levels 4 and 5 and the percentage was quite low
(3.4% and 1.1%). In contrast, the complexity level reflected in the informational writing
was relatively high, especially in the subcategory of reading reflections and analyses.
Students’ writing showed at least some degree of explanation from 1.3% (in the
subcategory of notes) to 47.4% (in the subcategory of reading reflections). Personal
diary was the only subcategory identified in personal writing and this was indexed at
the lower levels (one, two, and three).
13 | JOURNAL OF WRITING RESEARCH
Table 3. Levels of Epistemic Complexity across Types of Writing (Percentages)
3.1 Epistemic Complexity by Language Background, Grade Level, and
School Context
With regard to our second question (How does the level of epistemic complexity differ
by adolescents’ language background, grade level, and educational context?), we found
some differences evident in the levels of epistemic complexity by students’ language
backgrounds, grade levels, and school contexts although only school context was
found to be a statistically significant factor. It is important to note, as we will discuss
later in the limitations section, that tests for statistical significance are weak on such a
small sample.
Nonetheless, we found that although the mean level of epistemic complexity
reflected in NESs’ writing was higher than that in ELs’ writing, there was no statistically
significant difference in epistemic complexity among students of different language
backgrounds. In addition, while there was no statistical significance in epistemic
complexity by students’ grade levels, the levels of epistemic complexity rise slightly by
grade level from 1.49 at 6th grade to 2.17 at the 12th grade. Finally, the mean level of
epistemic complexity in the writing of students from California, Texas, and New York
was 1.98, 1.24, and 1.93, respectively and the mean level of complexity in writing of
students from the three states was statistically significant. Scheffé test was used to
further identify where the statistical significance was. Results showed that the difference
existed between the level of epistemic complexity in the writing of students from
California and Texas (p = 0.05; SE = 0.28). ANOVA test generated the same results of
the statistical significance with regard to students’ language background, grade level,
level 1 level 2 level 3 level 4 level 5 (nr of question items)
Mechanical
short answer 54.7 28.4 12.4 3.4 1.1 1,202
fill-in-blank exercises 100 0 0 0 0 261
symbolic 61.1 5.6 33.3 0.0 0 22
table 20 50 30 0 0 13
Informational
notes 58.5 22.7 17.5 0 1.3 140
lab reports 61.3 17.3 7.6 10.6 3.2 268
reading reflections 0 15.8 36.8 21.1 26.3 19
analyses 27.3 21.4 29.1 9.4 12.8 117
Personal
diary 50 0 50 0 0 4
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 14
and school context, when we excluded the five students with fewer than two pieces of
writing. These results can be seen in Table 4.
Table 4. Levels of Epistemic Complexity by Language Background, Grade Level, and School
Context
Student type
M
SD
N
F
p-value
Language Background
1.38
0.25
NES
1.85
0.8
18
EL
1.56
0.61
15
Grade level
1.22
0.32
6
1.49
0.47
7
8
1.63
0.84
10
10
1.64
0.73
9
12
2.17
0.73
7
State
4.44
0.02*
CA
1.98
0.86
11
TX
1.24
0.19
11
NY
1.94
0.72
11
Tota l
1.72
0.73
33
* p < 0.05
3.2 The Nature of Epistemic Complexity in Different Types of Writing
Our third research question explored the nature of writing that exhibited lower and
higher levels of epistemic complexity with particular interest in the kinds of writing in
which adolescents moved beyond explanation (levels one, two, and three) to
postulating causes, reasons and other relations or theories related to scientific
phenomena (levels four and five). Here we begin by describing some examples of the
types of mechanical writing we examined more closely (e.g., short answer and
symbolic writing) and follow with examples of informational writing (e.g., reading
reflections and labs) that illustrate contrasts in the types of writing associated with
higher and lower levels of epistemic complexity.
Mechanical writing: Sometimes, but not always simple
As noted earlier, even though the students participating in the study produced a variety
of writing in science classes, most of this work was mechanical in nature which did not
require going beyond explanation. While much of the mechanical writing in our
sample was indexed at the lowest level (level one) of epistemic complexity, and
generally, was indexed lower than informational writing, some types of mechanical
writing exhibited higher levels of epistemic complexity and we were particularly
interested in these. In some cases it appears that the higher level of epistemic
15 | JOURNAL OF WRITING RESEARCH
complexity in the writing was as a result of the type of task and in other cases was
related to the nature of the prompt.
Short answers and symbols: From lists to relationships.
Of the mechanical writing in our sample, the majority was in the form of textbook
chapter review questions that are usually taken directly from a course book or a pre-
printed teacher’s guide. These tasks require students to answer with one word or a short
phrase. The pieces of writing in our sample generally did not exhibit high levels of
epistemic complexity evidenced by the analysis of ideas in order to explain causal or
other relationships or theorizing about these relationships. An example of such a
textbook chapter review task found in the samples produced from a sixth grade
classroom in Texas asked students to define non-renewable resources. An English
learner in this class responded with the following phrase: “not replaced as it is being
used. Ex. fossil fuels, metal (recycle), uranium)”. This exemplifies epistemic complexity
of level one.
In contrast, some mechanical pieces such as short answer and symbolic writing,
qualified as levels two and three on our scale. In one such short answer example,
produced in an Honors Chemistry class and in response to the prompt: “concepts to
learn for this experiment”, a 10th grade native English speaker from California wrote
“Some solutions conduct electricity, only ionic compounds conduct electricity because
of the presence of ions.” The epistemic complexity here is level two showing evidence
of making a causal explanation between scientific concepts.
While 61.1% of writing coded as symbolic was indexed at level one in terms of
epistemic complexity, 33.3% of these kinds of writing were indexed at level three.
These level three symbolic writing samples demonstrated well-organized descriptive
information. For example, a sixth grade EL from Texas produced an elaborate diagram
of the geological dimensions and relationships between igneous and sedimentary rock
and how geological forces, such as volcanic activity, produce molten material (Figure
2). This piece was indexed at level three as the student did not only list terms, but also
connected these terms by providing details as to their relationships.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 16
Figure 2. Symbolic writing.
In another example, a native English speaking senior from California, drew directional
lines to articulate the relationship between phase changes as liquid moves through the
three states of matter. In this example, like the previous one, single words were used to
capture elements of the phase changes between states of matter.
Thus, not all mechanical writing in our sample was simple; in some cases short
answer and symbolic writing was related to higher complexity requiring the
reorganization of information. However, this writing fell short of providing evidence of
claims, counterclaims, and the provision of evidence to support claims.
Informational writing: Sometimes, but not always complex
The informational writing in our sample tended to be of higher epistemic complexity
than mechanical writing overall. Reading reflections, for example, more than any other
type of writing, were indexed at level five. Yet, some reading reflections were
associated with very low levels of epistemic complexity. Likewise, some lab reports
were associated with lower levels of complexity and others much higher. What is it,
precisely, about the nature of this writing that related to higher or lower levels of
17 | JOURNAL OF WRITING RESEARCH
epistemic complexity? In this section, we begin with descriptions of reading reflections
which is followed by lab reports to explore this question: As we describe examples of
these types of writing we focus on characteristics of those indexed at higher levels of
epistemic complexity and those that were not.
Reading reflections. Reading reflection writing prompts can take many forms such
as summaries and reports. The most important characteristic we noted in this regard is
that, unlike mechanical writing prompts for chapter review questions which tend to
have a restrictive response nature to them; that is, there is a correct answer the teacher
is looking for, reading reflection questions were often open-ended and provided the
student an opportunity to think about, reflect upon, and articulate their questions and
emerging understandings. In addition, they sometimes required students to engage in
analysis, synthesis, and summary. For example, Randy1, one of the 10th grade students
in California, enrolled in an Honors level Chemistry course, wrote analyses and
summaries of topical writing appearing in science or popular science journals. In these
Randy synthesized, analyzed, and reflected on the content. For example, in one writing
piece, entitled “Oderprints like fingerprints?” Randy not only summarizes the key
findings of the article, he also explains the scientific underpinnings justifying the use of
odorprinting to identify individuals (see Figure 3). He writes: “This is possible because
humans and other mammals have unique, genetically determined body odors called
odortypes.” Then, he applies his understanding as he elaborates on the potential
application of such scientific developments. This piece was categorized at level five,
not merely because it was more than a few sentences, but because Randy’s explanation
is explicit and postulates causes and reasons, and his reflection indicates application of
his understanding as he notes the possibility of odorprinting “being used to detect skin
diseases.”
1 All student names are pseudonyms.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 18
Figure 3. Summary
While it may be easy to suppose that since Randy, in the preceding example, is a high-
achieving student in an Honors Chemistry class, that very high levels of epistemic
complexity may only be associated with work in such classes and by students who
qualify to be in such classes, this was not always the case. We also had examples of ELs
in regular (non-Honors) classes who produced writing of equal complexity. For
instance, Shane, a tenth grade English learner from New York in a non-Honors science
class also produced writing indexed at level five. Shane was required to research a
chemical element and write a “Radio Biography” (a narrative script intended to be read
aloud to a non-scientific community). After citing several sources for his script, Shane
frames his script as an endorsement for the element Tungsten as a presidential
candidate. He argues his stances on key issues such as “Tungsten’s foreign policies are
also quite dreadful. He refuses to negotiate with oxygen or acids. If Tungsten is elected,
we will surely be prone to ozone attacks”. An excerpt of this piece is shown in Figure 4.
19 | JOURNAL OF WRITING RESEARCH
Figure 4. Report.
Shane’s one and a half page report synthesizes the key facts, but more importantly, his
explanation of what Tungsten can and won’t do, framed as a political candidate, shows
deep understanding of chemical relationships and causality, even going so far as
predicting what might become of this chemical element if placed in a political arena.
These two examples of reading reflections required students to analyze different
texts, synthesize information, and articulate their explanations with different audiences
and purposes in mind. The demands of these reading reflections are not simply “what
did you learn?”; These tasks, synthesizing popular science writing and reflecting on the
implications, and recasting a research report into a radio biography invited students to
fulfill many of the demands of the Common Core State Standards and the Next
Generation Science Standards described at the beginning of this article.
Lab reports.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 20
Like reading response, lab reports also varied in the levels of epistemic complexity. The
lab report, a hallmark of secondary science classwork, typically engages students in
articulating phases of the scientific method (e.g., identifying problems, hypothesizing,
articulating procedures, and making conclusions based on evidence). In one case,
Dave, a low achieving native English speaking student from New York, produced a
response to a lab assignment on water in a typed response. In this assignment and
working with another student, Dave summarized the conclusions, but also added an
“error analysis” section making inferences about why he and his lab partner may have
found different results in their data compared to other students (Figure 5). Dave and his
partner write: “In the lad [sic] an error could have occurred when we weighted the
items. We also could have not heated the crystals enough to get rid of all the water”.
This piece of writing was indexed at level three, relatively high, which would not have
been the case if the writing prompt itself did not call for identifying flaws in lab
procedures invoking what was discussed earlier as movement toward an evaluativist
stance where scientific knowledge is understood and explained as tentative and
contingent.
Figure 5. Lab.
Lab reports, though, did not always correlate with higher levels of epistemic
complexity. The majority of lab report writing in our sample was tightly scripted,
leaving little room for students to explain the phenomenon they were observing. Note
that 61.3% of all lab writing from our sample was indexed at level one and 17.3% was
categorized at level two. As evidence of this pattern, Shane, one of the ELs from New
21 | JOURNAL OF WRITING RESEARCH
York, engaged in a “Redox Reactions” lab and was prompted to burn copper over a
flame and describe the reaction that occurred. Shane writes “the copper quickly
burned, producing a yellow flame”. The entirety of the lab report asks students for short
descriptions of observations such as this one, but only asks students to construct a
chemical equation to represent the observed reaction, resulting in symbolic writing.
Nowhere throughout the lab is Shane asked to draw conclusions, recognize patterns, or
make inferences about his observation of natural phenomena and he does not do so.
The chemical equations demonstrate to the teacher grading the assignment that Shane
can construct balanced chemical equations, likely meeting a major learning objective
of this task, but Shane does not show evidence of making sense of his observations (i.e.,
reasoning) in light of scientific theories or concepts he has previously learned.
In sum, these examples were selected for their illustrative nature as they capture the
general pattern observed in student writing samples of a relationship between epistemic
complexity and both the task structure and qualities of writing prompts (whether
restrictive or not) in those tasks.
4. Limitations
As part of a complex larger study, the current study faced several limitations that are
important for readers to note in considering the findings and conclusions. First, the
sample was purposive and not intended to be representative of science writing among
adolescents in U.S. secondary schools. Generalizability was not a goal, but rather,
insight into the nature of science writing in what might be considered “better-case
scenario” schools (i.e., those with histories of exemplary writing performance). In
addition, one of the data points not of concern in the larger study was the writing
prompt or task description that foregrounded the students’ writing. In some cases, the
students’ writing followed these task descriptors and so this information was available
to us within the sample, however, this was oftentimes not the case. In any event, we
could not systematically analyze the tasks and prompts that would have provided
interesting insight into the relationships of the nature of tasks and prompts and the
qualities of the writing students produced in terms of epistemic complexity. We also
did not have access to what materials students may have used in preparation for their
writing and these materials may have influenced what they wrote such as in the case of
paraphrasing the language used in a source text. Therefore, whether a student actually
was engaging in a particular kind of reasoning or mimicking the reasoning of another
writer in their written work is unclear. Finally, the scale of epistemic complexity we
utilized is only one option that inevitably takes some things into account and not
others. So, for instance, although the example of the metaphor of Tungsten as a
political candidate was scored fairly high on the scale of epistemic complexity we used,
it might be viewed as a simple description using other measures.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 22
5. Discussion
We found that the types of writing secondary students in our study produced were
generally not of high levels of epistemic complexity. This is particularly true of English
learners, students at the lower grades (6th and 8th), and students in some school contexts
where there appears to be little emphasis on more complex writing in science
classrooms.
Although the total amount of writing English learners produced was more than that
of native English speakers, EL writing was largely mechanical and fell into levels one
and two while native English speaking students produced a higher percentage of
informational writing and overall their writing was more complex although these
differences were not statistically significant. Nevertheless, this finding suggests that
language background related to differences in the complexity of adolescents’ science
writing to some extent. This finding raises questions as to whether the root of this
contrast is in different opportunities for writing, English learners’ stage of language
development, or some combination of a variety of these and other factors.
Like the differences in epistemic complexity by language background, differences
by grade level were also not significant. However, the pattern that we noted showed an
increase in complexity evident in the writing of upper-classmen (i.e.,12th graders) in
comparison with students in lower grades. Indeed, one would expect that epistemic
complexity would be higher as students progressed in their schooling and this was the
case in our sample. Whether this was due to teachers’ expectations, students’ in-school
and out-of-school experiences, or some variety of other factors is unclear from our data.
The findings regarding differences by school contexts (e.g., students from New York
produced the highest percentage of informational writing and writing of higher
epistemic complexity in science overall than those in other states) may relate to the fact
that New York was the only state which had open-response questions on the high
stakes exit exam in science. This finding raises further questions as to the influence of
high stakes exams on teachers’ expectations for writing and instructional emphases in
secondary science classrooms.
Our analyses revealed that mechanical writing was associated with lower levels of
epistemic complexity in comparison to informational writing. Yet, while overall,
mechanical writing was associated with lower levels of epistemic complexity, it was
not exclusive of more complex explanations of relationships or theorizing about
scientific concepts. In fact, some of the student writing categorized as mechanical did
require some more complex articulations; albeit the examples of these were limited.
Likewise, we found that not all informational writing was complex; both reading
reflections and lab reports were associated with varying levels of complexity and in
some cases this appeared to relate to the task structure and in others the quality of the
prompt. In this regard, prompts that were characterized by invitations for students to
manipulate and make sense of content demonstrated higher levels of epistemic
complexity.
23 | JOURNAL OF WRITING RESEARCH
Although the results of this study are not generalizable to the writing experiences of all
secondary adolescents in the United States, they highlight the nature of writing in terms
of epistemic complexity among a students in a set of schools identified for exemplary
writing performance (potentially better-case scenario contexts) that nonetheless
produced writing at fairly low levels of epistemic complexity. The finding that the
sample showed evidence of a paucity of epistemically-complex science writing overall,
and a preponderance of writing at lower levels of epistemic complexity particularly in
the samples from English learners and middle-level students and those from contexts
where writing in science is not as strongly emphasized in high stakes exams hold
implications for future research and practice.
6. Conclusion
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 24
From a sociocultural perspective, our study highlights the myriad ways adolescents
were engaged in making sense of scientific phenomena in their writing, yet also
revealed that of all of the writing they produced, very little of it involved wrestling with
scientific problems as more expert scientists do. This finding draws attention to the
concerns outlined at the beginning of this article with regard to how well adolescents
are developing 21st Century skills and dispositions toward writing in the academic
disciplines and how well they are being prepared to engage in writing that aligns with
more rigorous standards such as the Common Core State Standards and Next
Generation Science Standards (Green, & Dixon, 1993; Jimenez-Aleixandre, Rodriguez,
& Duschl, 2000; Lemke, 1990; Rijlaarsdam, et al., 2006).
Although the Common Core State Standards and Next Generation Science
Standards place an emphasis on developing students’ capacities to make arguments and
understand the nature of evidence in the scientific domains, our study reveals that
many of the students in even historically better schools in terms of writing instruction
may not achieve these standards in their writing. The reasons for this are likely multiple
including, as other researchers have found, (Owen, 2001; Tsai, 2002), that oftentimes
teachers’ beliefs about the nature of science as it might be revealed in discourse
including writing - is reproduced rather than constructed. However, as standards and
standardized testing requirements shift to a stronger emphasis on more advanced
disciplinary discourse, secondary science teachers may change their beliefs about the
value of writing and may be offered guidance and support in developing their
understandings of the epistemological foundations of scientific knowledge and how
these are expressed in different kinds of writing.
The building blocks may already exist for this transformation. Our analyses revealed
that even some forms of mechanical writing such as short answers, that are already part
of many science teachers’ repertoires, may facilitate adolescents’ development in
argumentation and explanation if they are well-prompted and scaffolded. As Langer
and Applebee noted in their study decades ago, “writing tasks differ in the breadth of
information drawn upon and the depth of processing of that information that they
invoke” (1987, p. 131). So, while we noted that reading response and lab report tasks
were associated with higher levels of complexity overall, we also identified those that
were not. We also found writing samples that qualified as mechancial (e.g., short
answer and symbolic) in which students engaged in articulating detailed
understandings going well beyond recall of facts and others that were of higher
epistemic complexity. In the end, the nature of the task and prompt matter a great deal
as we have highlighted in our illustrative examples, and teachers can learn how to use
writing to achieve different aims if they also understand how to construct the tasks and
prompts carefully.
Our earlier discussion of the epistemology of science and its relationship to
epistemic complexity in writing may shed some insight into questions teachers may
have as to why certain writing tasks and prompts might be proposed in particular ways.
Drawing on Kuhn & Crowell’s (2011) work in the teaching of dialogic argumentation as
25 | JOURNAL OF WRITING RESEARCH
a means for students to internalize more mature understandings of the epistemology of
science, we glean some guidance as to how teachers might frame writing tasks as a
dialogic endeavor. Like the example “odorprints” described earlier, a science writing
pedagogy that engaged adolescents in dialogic argumentation including explaining,
justifying, and identifying causal relationships with regard to some scientific
phenomenon, followed by a task prompting them to express these ideas and their
implications in relationship to existing scientific knowledge and theories, might be
fruitful. Such an instructional approach may help students begin to grasp the
epistemological underpinnings of science content while also developing their
disciplinary writing competence. Secondary school teachers of science in an era of the
CCSS and NGSS and guided by an emphasis on the development of 21st Century
dispositions and skills might consider the kinds of writing tasks and prompts associated
with higher levels of epistemic complexity such as these as essential components of
adolescents’ preparation for the workplace or post-secondary study. As we noted in our
review of literature, if one of the purposes of writing in science is to “foster deep
thinking about science” (Keys, 1998), then secondary teachers need to think carefully
about the intended outcomes of assigning tasks that constrain adolescents to solely
recall of facts.
This study, informed and complemented by others (Rijlaarsdam, Couzijn, Janssen,
Braaksma, & Kieft, 2006; Rivard, 1994; Tsai, 2002; Wellington & Osborne, 2001),
contributes to the research literature an account of the kinds of writing tasks a variety of
students from different secondary school contexts produce. It also offers a potentially
useful approach to the analysis of science writing, the results of which might provide
some guidance to secondary science teachers in contemplating the qualities of writing
tasks as they relate to intended outcomes. If, as we have argued, one of the values of
writing in science is to prompt deep understanding of content including the
epistemological underpinnings of scientific knowledge, and ultimately build
dispositions and skills to engage in and with the scientific community as young adults,
further investigation beyond this study is needed. Studies exploring the relationships
between the qualities of tasks and prompts and levels of epistemic complexity in a
broader sample of adolescents’ writing may help build a stronger and deeper
conceptualization of science writing development and its pedagogical applications.
References NOTE please add DOI nrs, easily to get from:
http://www.crossref.org/SimpleTextQuery
Author. (2013) http://www.crossref.org/SimpleTextQuery
Achieve. (2012). Next Generation Science Standards, Retrieved from http://www.nextgeneration
science.org
Applebee, A. N. (1984). Writing and reasoning. Review of Educational Research, 54(4), 577-596.
Applebee, A. N. (1981). Writing in the secondary school: English and the content areas (Research
Monograph 21). Urbana, IL: National Council of Teachers of English.
Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition. New Jersey: L.
Erlbaum Associates.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 26
Brem, S. K., & Rips, L. J. (2000). Explanation and evidence in informal argument. Cognitive
Science, 24(4), 573-604.
Bricker, L., & Bell, P. (2009). Conceptualizations of argumentation from science studies and the
learning sciences and their implications for the practices of science education. Science
Education, 92(3), 473-498.
Common Core State Standards Initiative. (2010). Common core state standards for mathematics.
Retrieved from http://www.corestandards.org/the-standards
Chi, M. T., Glaser, R., & Rees, E. (1981). Expertise in problem solving. In R. Sternberg (Ed.),
Advances in the psychology of human intelligence (pp. 7-76). NJ: Erlbaum.
Chuy, M., Scardamalia, M., & Bereiter, C. (2012). Development of ideational writing through
knowledge building: Theoretical and empirical bases. In E. Grigorenko, E. Mambrino & D.
Preiss (Eds.), Writing: A mosaic of new perspectives. New York, NY: Taylor & Francis.
Duschl, R. (2008). Science education in three-part harmony: Balancing conceptual, epistemic, and
social learning goals. Review of Research in Education, 32(1), 268-291.
Fellows, N. J. (1994). A window into thinking: Using student writing to understand conceptual
change in science learning. Journal of Research in Science Teaching, 31(9), 985-1001.
Green, J. L., & Dixon, C. N. (1993). Talking knowledge into being: Discursive and social practices
in classrooms. Linguistics and Education, 5(3), 231-239.
Hakkarainen, K. (2003). Progressive inquiry in a computersupported biology class. Journal
of Research in Science Teaching, 40(10), 1072-1088.
Hakkarainen, K. P. J. (1998). Epistemology of scientific inquiry and computer-supported
collaborative learning. Retrieved from Dissertation Abstracts International.
Hand, B., Lawrence, C., & Yore, L. D. (1999). A writing in science framework designed to enhance
science literacy. International Journal of Science Education, 21(10), 1021-1035.
Hand, B., Wallace, C. W., & Yang, E. M. (2004). Using a science writing heuristic to
enhance learning outcomes from laboratory activities in seventhgrade science:
quantitative and qualitative aspects. International Journal of Science Education,
26(2), 131-149.
Hand, B., & Prain, V. (2002). Teachers implementing writingtolearn strategies in junior
secondary science: A case study. Science Education, 86(6), 737-755.
Hempel, C. G. (1965). Aspects of scientic explanation and other essays in the philosophy of
science. New York: The Free Press.
Jimenez-Aleixandre, M. P., Rodriguez, A. B., & Duschl, R. A. (2000). "Doing the lesson" or" doing
science": Argument in high school genetics. Science Education, 84(6), 757-792.
Kelly, G. J., & Bazerman, C. (2003). How students argue scientific claims: A rhetoricalsemantic
analysis. Applied Linguistics, 24(1), 28-55.
Kelly, G. J., Chen, C., & Crawford, T. (1998). Methodological considerations for studying science-
in-the-making in educational settings. Research in Science Education, 28(1), 23-49.
Kelly, G. J., & Crawford, T. (1997). An ethnographic investigation of the discourse processes of
school science. Science Education, 81(5), 533-559.
Kelly, G. J., & Takao, A. (2002). Epistemic levels in argument: An analysis of university
oceanography students' use of evidence in writing. Science Education, 86(3), 314-342.
Keys, C. W. (1999). Revitalizing instruction in scientific genres: Connecting knowledge production
with writing to learn in science. Science Education, 83(2), 115-130.
Konstantinidou, A., & Macagno, F. (2013). Understanding Students' Reasoning: Argumentation
Schemes as an Interpretation Method in Science Education. Science & Education, 22(5), 1069-
1087.
Kuhn, D. (2010). Teaching and learning science as argument. Science Education, 94(5), 810-824.
Kuhn, D. (1993). Science as argument: Implications for teaching and learning scientific thinking.
Science Education, 77(3), 319-337.
Kuhn, D., Iordanou, K., Pease, M., & Wirkala, C. (2008). Beyond control of variables: What needs
to develop to achieve skilled scientific thinking?. Cognitive Development, 23(4), 435-451.
27 | JOURNAL OF WRITING RESEARCH
Kuhn, D., & Crowell, A. (2011). Dialogic Argumentation as a Vehicle for Developing Young
Adolescents’ Thinking. Psychological Science, 22(4), 545-552.
Kuhn, D., Cheney, R., & Weinstock, M. (2000). The development of epistemological
understanding. Cognitive Development, 15(3), 309-328.
Langer, J. A., & Applebee, A. N. (1987). How writing shapes thinking: A study of teaching
and learning. NCTE Research Report No. 22. National Council of Teachers of
English, Urbana, IL.
Lemke, J. L. (1990). Talking science: Language, learning, and values. NJ: Ablex Publishing
Corporation.
MacArthur, C. A., Graham, S., & Fitzgerald, J. (Eds.). (2006). Handbook of writing research.
New York: Guilford Press.
Metz, K. E. (2008). Narrowing the gulf between the practices of science and the elementary school
science classroom. The Elementary School Journal, 109(2), 138-161.
Moje, E. B. (2011). Developing disciplinary discourses, literacies, and identities: What's
knowledge got to do with it?. In L. Bonilla & Englander (Eds.), Discourses and identities in
contexts of educational change (pp. 49-74). New York: Peter Lang.
National Governors Association (NGA) Center for Best Practices & Council of Chief State School
Officers. (2010). Common Core State Standards for English language arts and literacy in
history/social studies, science, and technical subjects. Retrieved from
www.corestandards.org/assets/CCSSI_ELA%20Standards.pdf
Norris, S. P., & Phillips, L. M. (2003). How literacy in its fundamental sense is central to scientific
literacy. Science Education, 87(2), 224-240.
Osborne, J., Erduran, S., & Simon, S. (2004). Enhancing the quality of argumentation in school
science. Journal of Research in Science Teaching, 41(10), 9941020.
O'Neill, D. K. (2001). Knowing when you've brought them in: Scientific genre knowledge and
communities of practice. The Journal of the Learning Sciences, 10(3), 223-264.
Owens, C. V. (2001). Teachers’ responses to science writing (Research Report No. ED 457 157).
Retrieved from Educational Resources Information Center website:
http://files.eric.ed.gov/fulltext/ED457157.pdf
Pearson, P. D., Moje, E., & Greenleaf, C. (2010). Literacy and science: Each in the service of the
other. Science, 328(23), 459-463.
Porter, A. C., Kirst, M. W., Osthoff, E., Smithson, J. L., & Schneider, S. A. (1994). Reform of high
school mathematics and science and opportunity to learn. Consortium for Policy Research in
Education, 13, 1-8.
Rijlaarsdam, G., Couzijn, M., Janssen, T., Braaksma, M., & Kieft, M. (2006). Writing experiment
manuals in science education: The impact of writing, genre, and audience. International
Journal of Science Education, 28(2-3), 203-233.
Rivard, L. O. P. (1994). A review of writing to learn in science: Implications for practice and
research. Journal of Research in Science Teaching, 31(9), 969-983.
Salmon, W. (1984). Scientific explanation and the causal structure of the world. NJ: Princeton
University Press.
Salmon, W. C. (1978). Why ask," Why?"? An inquiry concerning scientific explanation. In
Proceedings and addresses of the American Philosophical Association, 51(6), 683-705.
Sandoval, W. A. (2003). Conceptual and epistemic aspects of students' scientific explanations. The
Journal of the Learning Sciences, 12(1), 5-51.
Singer, S. R., Hilton, M. L., & Schweingruber, H. A. (Eds.). (2006). America's lab report:
Investigations in high school science. National Academies Press.
Tsai, C. C. (2002). Nested epistemologies: science teachers' beliefs of teaching, learning and
science. International journal of science education, 24(8), 771-783.
Wellington, J. J., Osborne, J., & Wellington, J. J. (2001). Language and literacy in science
education. Buckingham: Open University Press.
Yin, R. K. (2005). Case study research: Design and methods (3rd ed.). Thousand Oaks: Sage.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 28
Zhang, J., Scardamalia, M., Lamon, M., Messina, R., & Reeve, R. (2007). Socio-cognitive dynamics
of knowledge building in the work of 9-and 10-year-olds. Educational Technology Research
and Development, 55(2), 117-145.
29 | JOURNAL OF WRITING RESEARCH
Appendix A. School Demographics
Stat
e
School
Size
Grade
Span
% F/R
L
%
ELL
% African-
American
%
Hispanic
%
White
% Asian or Native
Hawaiian/
Other Pacific Islander
% American Indian/Alaska
Native
District-wide total per pupil
$ expenditure
CA
Montebello
1,66
4
5 to 8
83
28
0.2
96.5
2
1.1
0.1
$8,764
CA
John Adams
977
6 to 8
44
16
10.2
50.2
33.1
4.4
0.1
$10,130
CA
King Drew
1,68
0
9 to 12
66
3
59.5
37.8
0.3
0.9
0.2
$10,590
NY
Albert
Leonard
1,19
5
6 to 8
25
1
29
22
45
5
0
19,356
NY
Port Chester
794
6 to 8
43
12
9
72
18
1
0
17,046
NY
Batavia
763
9 to 12
34
0
8
2
87
2
1
$16,928
NY
New Paltz
803
9 to 12
14
0
7
6
84
3
0
$18,016
TX
Grisham
657
6 to 8
19
4
6.7
17.7
59.4
16.2
0
$7,191
TX
Spring
Branch
763
6 to 8
20
3
2.2
25
70.5
2.2
0
$6,926
TX
McCallum
1,71
8
9 to 12
35
5
21.6
29.8
46.3
2
0.3
$8,141
TX
Round Rock
2,64
8
9 to 12
24
5
10.7
26.6
57.2
5.1
0.4
$7,191
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 30
Appendix B. Categories of Writing
The categories of function refer to the way the language is used in a piece of writing.
Below are four main categories of function and a brief description of each based on the
work of Applebee (1981).
Writing without Composing (or Mechanical Uses of Writing): Tasks which require
written responses but that do not require the writer to organize text segments of more
than a paragraph length. Subcategories include: Multiple-choice exercises, fill-in-the-
blank exercises, short answer exercises, transcription from written material (copying) or
oral material (dictation), translation, symbolic expression (diagrams, graphs).
Informational Writing: Writing which focuses on the sharing of information or opinions
with others. This includes the wide variety of forms of expository writing, ranging from
simple reports about specific events to highly abstract, theoretical arguments. It also
includes writing where the attempt to persuade overrides all other purposes (as in
advertisements or propaganda), and regulative writing (e.g., laws or school rules).
Subcategories include: Note taking, record, report, summary, analysis, theory,
persuasive essay.
Personal Writing: Writing that is embedded within a context of shared, familiar
concerns. The audience for such writing is usually the self or a very close friend; the
function is to explore new ideas and experiences simply to sort them out, rather than to
make a specific point. Gossip in spoken language illustrates the general category; in
school writing, this use occurs mostly in journals or “learning logs” where new ideas
are explored for the writer’s own benefit. Subcategories include: Journal, diary, notes,
personal letters.
Imaginative Writing: Writing within any of the various literary genres. Subcategories
include: Stories, poems, play scripts.
31 | JOURNAL OF WRITING RESEARCH
Appendix C. Coding Scheme for Epistemic Complexity
Level 1: Separated pieces of facts. A rating of 1 was assigned to students’ writing
statement if it was transformed to a concept map with one entity or a group of entities
without any relationship. Entities might have a descriptive or multiple descriptives, but
no relationship existed between different entities. For example:
Some related animals are, Sponges, Venuses Flower basket, Portuges[sic] man of
war, Sea Anomes [sic], Jelly Fish and Hydra.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 32
Level 2: Partially-organized Facts. A rating of 2 was given to ideas that represented
loosely connected pieces of factual information. Visually, two or multiple entities had
only one layer of relationship within or among each other. For example:
Dermal bones form in subcutaneous membranes. They’re mostly composed of
cancellouse bone and parts of irregular bones.
33 | JOURNAL OF WRITING RESEARCH
Level 3: Well-organized Facts. A rating of 3 was assigned to writing statements in which
factual information was introduced in a well-organized way. It was represented by two
or multiple entities with multiple layers of a relationship within or among each other on
the concept maps.
Endochondral bones form from cartilage pegs in the embryo. They usually produce
long bones, and parts or irregular and short bones. Endochondral bones have primary
and secondary ossification centers, and a region that produces the bone collar.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 34
Level 4: Partial explanation. A rating of 4 was assigned to ideas that represented some
characteristics of an explanation but the content of the explanation was limited or only
partially articulated. It was transformed to concept maps as entities with a cause/effect
relationship, but the logic sequence was incomplete, meaning some essential entities or
relationships were missed. For example:
Osteonecrosis is death of bone caused by loss of blood flow to the oseteons.
35 | JOURNAL OF WRITING RESEARCH
Level 5: Well-organized explanation. A rating of 5 was given to writing statements for
which a relatively well elaborated explanation was provided. On the concept maps,
two or multiple entities had a cause/effect relationship with a complete logic sequence
containing all essential nodes. For example:
A drug described as antagonistic would decrease the normal neurotransmitter
response. The mechanism of such a drug if it were to exert it effects on the postsynaptic
side of the synaptic cleft would result in less sensitivity to neurotransmitters.
WILCOX, YU & NACHOWITZ EPISTEMIC COMPLEXITY IN ADOLESCENT SCIENCE WRITING | 36
Appendix D. Levels of Epistemic Complexity by Language Background, Grade Level,
and School Context
Student type
n
M(SD)
F-value
p
Language Background
1.29
0.27
NES
18
1.85 (0.8)
EL
15
1.56 (0.61)
Grade level
1.22
0.32
6
7
1.49 (0.47)
8
10
1.64 (0.83)
10
9
1.64 (0.73)
12
7
2.17 (0.73)
State
4.44
0.02*
CA
11
1.98 (0.86)
TX
11
1.24 (0.19)
NY
11
1.94 (0.72)
Tota l
33
1.72 (0.73)
Note. * p < 0.05
... To respond to our second research question, regarding students' modes of participation, two case study students' written products were examined for evidence of the influence of key events through intertextual tracings (Newell et al., 2021;Wynhoff Olsen et al., 2018). In addition, epistemic complexity of the written products (Wilcox et al., 2015) provided a way to characterize how the writing functioned as an instrument for knowledge representation. This analysis was based on a scale from simple descriptions of scientific phenomena to increasingly more complex explanations of relationships of phenomena and analyses of phenomena that include arguments. ...
... Delucca] asked us to write based on the questions he gave us." We used two different approaches to interpret the report: (a) intertextual connections (Bazerman, 2004) to consider the multiple sources drawn upon during the unit, and (b) epistemic complexity (Wilcox et al., 2015) to study the extent to which the lab reports included arguments to explain phenomena and postulate causes, reasons, and other relations or theories related to scientific phenomena. Below is William's lab report with each message unit numbered for our reference purposes: ...
Article
Full-text available
This study explores one teacher's instructional method for teaching life sciences using argumentation and argumentative writing rather than simple templates for writing claims and evidence. The microethnographic discourse analytic case study reported here included the teacher and 26 “advanced” eighth-grade students in a suburban middle school. Nine consecutive video-recorded lessons and related data were analyzed, focusing on how the teacher and students constructed the theory of evolution during instructional conversations about evidence and reasoning and about the content of students’ written arguments on the theory. The teacher created a context in which students developed arguments with teacher support to ensure that they were learning to use argumentation as a heuristic to understand concepts and to engage in argumentative practice central to doing science. The modes of participation of two case study students are contrasted to explore two different trajectories and to examine particular cases of writing a lab report.
... Otro tanto cabe afi rmar en lo relativo al aprendizaje en la educa- ción en ciencias ( Halliday and Martin, 1993;Sanmartí, 1996Sanmartí, , 2007; Wäschle, K., Gebhardt, A., Oberbusch, E.-M, and Nückles, M., Wilcox, Yu, and Nachowitz, 2015;Roni y Carlino, 2017). Estos autores postulan que el lenguaje es una herramienta fundamental que se usa para clasifi car, descomponer y analizar, explicar, y exponer las ave- riguaciones a que llega un científi co. ...
... Por ello, la escritura es principalmente un recurso de aprendizaje ya que gran parte del conocimiento que se produce se genera y se difun- de a través de textos que se escriben y se leen (Rose and Martin, 2012;Camps y Castelló, 2013). En este sentido, la escritura cumple una fun- ción epistémica ya que es una herramienta idónea para aprender, en cuanto que el proceso de producción textual favorece la transformación de las estructuras cognitivas del escritor ( Balgopal and Wallace, 2013;Bereiter and Scardamalia, 1987;Klein and Rose 2010;Wilcox, Yu, and Nachowitz, 2015), a la vez que es un instrumento de representación del conocimiento (Chuy, Scardamalia and Bereiter, 2012). ...
Book
Full-text available
Este trabajo recoge aportaciones muy valiosas generadas como resultado del XXI Encuentro Internacional de Investigación Educativa y II Encuentro Internacional de Comunicación Pública de la Investigación Educativa: Lectura, escritura y matemáticas, textos y contextos. Literacidades en el mundo actual, en el marco de la Feria Internacional del Libro en su edición número 31. El solo hecho de que se realice este encuentro ya por más de dos décadas habla de que la investigación académica no puede estar separada de la difusión general de la cultura. La universidad y las editoriales, entre otras de las instituciones encargadas de generar conocimiento y ofrecerlo al mundo, caminan juntas en medio de un paradigma donde parece lidiarse una batalla entre las librerías e Internet; entre la biblioteca y la computadora; entre las fuentes impresas y las electró-nicas, tanto en lo que se refiere a la búsqueda de ese conoci-miento como a su publicación, y es justamente la necesidad social de leer lo que justifica la existencia de esta condición. Con referencia a su estructura este libro contiene 9 capitu-los, los cuales incluyen ensayos, reportes de investigación y revisión de estudios diversos. En ellos se plantea la relación entre la lectura, la escritura y las matemáticas como objetos de aprendizaje en distintos niveles educativos, así como las diver-sas nociones sobre los conceptos de literacidad y alfabetización.
... Otro tanto cabe afi rmar en lo relativo al aprendizaje en la educa- ción en ciencias ( Halliday and Martin, 1993;Sanmartí, 1996Sanmartí, , 2007; Wäschle, K., Gebhardt, A., Oberbusch, E.-M, and Nückles, M., Wilcox, Yu, and Nachowitz, 2015;Roni y Carlino, 2017). Estos autores postulan que el lenguaje es una herramienta fundamental que se usa para clasifi car, descomponer y analizar, explicar, y exponer las ave- riguaciones a que llega un científi co. ...
... Por ello, la escritura es principalmente un recurso de aprendizaje ya que gran parte del conocimiento que se produce se genera y se difun- de a través de textos que se escriben y se leen (Rose and Martin, 2012;Camps y Castelló, 2013). En este sentido, la escritura cumple una fun- ción epistémica ya que es una herramienta idónea para aprender, en cuanto que el proceso de producción textual favorece la transformación de las estructuras cognitivas del escritor ( Balgopal and Wallace, 2013;Bereiter and Scardamalia, 1987;Klein and Rose 2010;Wilcox, Yu, and Nachowitz, 2015), a la vez que es un instrumento de representación del conocimiento (Chuy, Scardamalia and Bereiter, 2012). ...
Chapter
Full-text available
Este capítulo analiza distintos enfoques de investigación y sus resultados respecto de concepciones de escritura académica y aprendizaje en estudiantes universitarios y argumenta la relevancia de desarrollar investigación científica en México sobre dicha temática.
... Measuring epistemic complexity is one method for analyzing depth of student understanding from written products (see, for example, Hakkarainen, 2003;Kelly & Takao, 2002;Zhang, Scardamalia, Lamon, Messina, & Reeve, 2006). The categories of epistemic complexity used in this analysis are slightly modified from these earlier studies and taken directly from recent work studying the epistemic complexity of students' science writing (Wilcox, Yu, & Nachowitz, 2015). The author coded randomly selected, mathematical writing samples identified as short-answer exercises (n = 294) on a five-point scale moving from unsupported statements (low complexity), through explanation and description, to the highest level evidencing argumentation (claims and counterclaims). ...
Article
Using a national sample of students’ writing from the National Study of Writing Instruction, this study examines the current state of writing in the mathematics classroom and the extent to which enactments align with the intentions established in the research base and curricula articulated by NCTM and the CCSS–M. Over 3,000 writing assignments from 138 students in grades, 6, 8, 10, and 12 were analyzed for functions of writing and epistemic complexity—the degree to which they explained or justified mathematical understanding. Additionally, shortanswer assignments were analyzed qualitatively for patterns in teacher-assigned prompts and student responses. Analyses reveal enactments of writing in the mathematics classroom do not align with the intent articulated by the CCSS-M. Mirroring techniques from the literacy field, a method for scaffolding students’ improved ability to write about conceptual understanding in mathematics is discussed.
... To classify these writing to learn strategies according to the functions (i.e., mechanical, informative, personal, and imaginative) identified by Applebee (1981), we obtained a scoring manual used for this purpose in a study by Wilcox et al. (2015). We then sorted the 43 strategies into the four functions, using the definitions and examples provided in the manual. ...
Article
Full-text available
A random sample of middle school teachers (grades 6–9) from across the United States was surveyed about their use of writing to support students’ learning. The selection process was stratified so there were an equal number of English language arts, social studies, and science teachers. More than one-half of the teachers reported applying 15 or more writing to learn strategies at least once a month or more often. The most commonly used writing to learn strategies were writing short answers to questions, note taking for reading, note taking while listening, and completing worksheets. While teachers reported using a variety of writing to learn strategies, most of them indicated they received minimal or no formal preparation in college on how to use writing to learn strategies to support student learning, less than one-half of teachers directly taught students how to use the writing to learn strategies commonly assigned, and the most commonly used writing to learn strategies did not require students to think deeply about the material they were learning. We further found that teachers’ reported use of writing to learn strategies was related to their preparedness and the composition of their classroom in terms of above and below average writers, English Language Learners, and students with disabilities.
Article
This paper highlights the perspectives of first-generation students of color in their first year of college, and the ways in which they exercised agency in their writing. Framed by definitions of agency as mediated action that creates meaning, the paper reports on qualitative data collected from a summer writing program for first-generation students and students of color, and from writing samples and follow-up interviews with six students who participated in the summer program. Findings suggest that students in their first year of college leveraged their social and discoursal identities to offer new ways of understanding an issue. They also wrote using a translingual approach, integrating different discourses and forms of knowledge, and challenging views of academic writing as monolithic. The findings also suggest the link between awareness and action, meaning that what and how students wrote were informed by their awareness of writing and awareness of themselves as writers and cultural beings. The study’s findings have implications for advancing more nuanced views of agency and academic literacies, and redesigning writing instruction at the high school and college level.
Article
Full-text available
In most assessments of students’ argumentive writing and in most research on the topic, students write on topics for which they have no specific prior preparation. We examined development in the argumentive writing urban middle school students did as part of a two-year dialogic-based intervention in which students engaged deeply with a series of topics and participated in various kinds of electronic and verbal dialogic activities related to the topic. Students’ achievements exceeded those typical of the extemporaneous expository writing of middle school students. The gradual developments observed over time included ones in (a) addressing and seeking to weaken the opposing position, (b) identifying weaknesses in a favored position and strengths in an opposing one, (c) connecting and integrating opposing arguments, and (d) using evidence to weaken as well as support claims. In addition to identifying a trajectory of what develops in the development of argumentive writing, these analyses support the claim of dialogic argumentation as a productive bridge to individual argumentive writing and highlight the contribution of deep engagement with the topic in enhancing writing.
Article
Full-text available
Research on writing to learn is conceptually rich and pedagogically important. This special issue contributes to our growing knowledge about the variables that mediate and moderate the effects of writing on learning. One group of studies addresses individual writing; it adds to the growing evidence for the effects of several moderators, including cognitive, metacognitive, and personal utility prompts in journal writing (Wäschle et al.); and discipline-specific prompts for argument writing (Van Drie et al.). The individual-level studies also suggest moderator roles for discipline-general strategies of argumentation (Smirnova); and open-ended informational writing assignments (Wilcox et al.). Additionally, the Wäschle et al. study provides the strongest evidence for a specific sequence of mediation: Prompted journal writing to comprehension to interest to critical reflection. The second set of articles focusses on collaborative writing (Corcelles & Castelló; Ortoleva and Bétrancourt). It suggests moderating roles for discipline-specific analytic strategies, and peer interaction; and suggests mediating roles for exploratory discourse and group regulation. Further experimental research on collaborative writing is needed to conclusively test hypotheses about specific mediator and moderator variables. The studies in this special issue, like many recent studies, are incommensurable with some influential theories of writin g to learn. Rather, individual and collaborative studies converge to suggest that writing to learn may be conceptualized as a guided process of cognition and self-regulation.
Article
Full-text available
This brief concerns the nature of the high school mathematics and science curriculum in the United States. It draws from a large study which documented instructional practices and content using novel methodologies. This research approach is a promising step toward the development of indicators of opportunity to learn. The study also provides encouraging news about the effects of increased standards in math and science - they did not result in a watering down of the curriculum. However, practice in the schools studied is a far cry from the ambitious goals for math and science instruction now being developed by the profession.
Book
Laboratory experiences as a part of most U.S. high school science curricula have been taken for granted for decades, but they have rarely been carefully examined. What do they contribute to science learning? What can they contribute to science learning? What is the current status of labs in our nation's high schools as a context for learning science? This book looks at a range of questions about how laboratory experiences fit into U.S. high schools: What is effective laboratory teaching? What does research tell us about learning in high school science labs? How should student learning in laboratory experiences be assessed? Do all student have access to laboratory experiences? What changes need to be made to improve laboratory experiences for high school students? How can school organization contribute to effective laboratory teaching? With increased attention to the U.S. education system and student outcomes, no part of the high school curriculum should escape scrutiny. This timely book investigates factors that influence a high school laboratory experience, looking closely at what currently takes place and what the goals of those experiences are and should be. Science educators, school administrators, policy makers, and parents will all benefit from a better understanding of the need for laboratory experiences to be an integral part of the science curriculum and how that can be accomplished. © 2006 by the National Academy of Sciences. All rights reserved.
Article
A substantial body of evidence shows that people tend to rely too heavily on explanations when trying to justify an opinion. Some research suggests these errors may arise from an inability to distinguish between explanations and the evidence that bears upon them. We examine an alternative account, that many people do distinguish between explanations and evidence, but rely more heavily on unsubstantiated explanations when evidence is scarce or absent. We examine the philosophical and psychological distinctions between explanation and evidence, and show that participants use explanations as a substitute for missing evidence. Experiment 1 replicates the results of other researchers, but further shows that participants generate more evidence when they are not constrained by their lack of data. Merely mentioning a source of data can alter both their evaluation (Experiment 2) and their production (Experiment 3) of explanations and evidence. In Experiment 4, we show that participants can explicitly consider the availability of evidence and other pragmatic factors when evaluating arguments. Finally, we consider the implications of using explanations to replace missing evidence as a strategy in argument.
Article
[In ‘The Philosophy of Hans Reichenbach’ (above), I remarked upon the fertility of his posthumous work, The Direction of Time, especially its fourth chapter, which deals with the causal and statistical relations among macrophenomena. I emphasized the value of his ideas in connection with the explication of scientific explanation — in particular, their potential utility for purposes of developing a full-blown theory of causal explanation. Since that article was written, I have been devoting considerable attention to just such issues, and I feel that my hopes have been —at least to some extent — vindicated.
Book
Presents a collection of essays discussing the theories and models of writing research.