Braaten 2011

SCIENCE STUDIES AND SCIENCE EDUCATION
Sibel Erduran and John L. Rudolph, Section Coeditors
Working Toward a Stronger

Conceptualization of Scientific
Explanation for Science Education
MELISSA BRAATEN
Department of Curriculum and Instruction, University of Wisconsin,
Madison, WI 53706-1795, USA
MARK WINDSCHITL
Department of Curriculum and Instruction, University of Washington,
Seattle, WA 98195-3600, USA
Received 11 May 2010; revised 11 November 2010; accepted 29 November 2010
DOI 10.1002/sce.20449
Published online 23 May 2011 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: Scientific explanation plays a central role in science education reform doc-
uments, including the Benchmarks for Science Literacy, the National Science Education
Standards, and the recent research report, Taking Science to School. While scientific ex-
planation receives significant emphases in these documents, there is little discussion or
consensus within the science education community about the nature of explanation itself.
However, debates about scientific explanation have been a mainstay for philosophers of
science for decades. We argue that a more clearly articulated conceptualization of scientific
explanation for science education is necessary for making the vision of science education
reform a reality. In this essay, we use major philosophical theories of scientific explana-
tion as lenses to examine how the science education community has constructed the idea of
Correspondence to: Melissa Braaten; e-mail: mbraaten@uw.edu

Contract grant sponsor: Teachers for a New Era Project was sponsored by the Carnegie Corporation,
Annenberg Foundation, and the Rockefeller Foundation.
Contract grant sponsor: National Science Foundation.
Contract grant number: DRL-0822016.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the
authors and do not necessarily reflect the views of the funding organizations.

C 2011 Wiley Periodicals, Inc.
640 BRAATEN AND WINDSCHITL
explanation. We also examine instructional practice in school science settings, including our
own classrooms, where teachers and students are working to explain natural phenomena.
Using these examples, we offer suggestions for preparing both educators and young learners
to engage in explanatory discourses that are reasonably accountable to authentic epistemic
practice in science. C 2011 Wiley Periodicals, Inc. Sci Ed 95:639 – 669, 2011
INTRODUCTION
Science educators engaged in reforming science education envision classrooms where
students and teachers focus on key ideas and practices in science especially through the
construction and understanding of scientific explanations (American Association for the
Advancement of Science [AAAS], 1993; Harrison & Treagust, 2000; Millar & Osborne,
1998; Mortimer & Scott, 2003; National Research Council [NRC], 1996, 2007; Osborne
& Dillon, 2008; Treagust & Harrison, 2000). This press for explanation is believed to be
crucial for engaging students in some of the most central epistemic practices of science and
for developing robust understanding (Chinn & Malhotra, 2002; Duschl & Grandy, 2008;
Driver, Asoko, Leach, Mortimer, & Scott, 1994; Windschitl, 2008). Unfortunately, within
the field of science education “scientific explanation” is not well defined for educational
researchers, classroom teachers, and ultimately for science students. While there are many
scholars in science education working to support teachers’ and students’ efforts at pressing
for scientific explanations, the projects are often undertheorized with regard to scientific
explanation. The lack of clear vision and definition about the substance and function of
scientific explanations leaves teacher educators like us and the early career teachers with
whom we work struggling to support students’ work with scientific explanations.
The current practice of many experienced teachers tends to focus on students accumulat-
ing and repeating descriptive information about natural phenomena and engaging students
in observational exercises or rudimentary experiments without pressing students toward
scientific explanations of phenomena (Banilower, Smith, Weiss, & Pasley, 2006; NRC,
2007; Osborne & Dillon, 2008; Roth & Garnier, 2007). In order for science educators to
be able to encourage young students to make shifts from descriptions to explanations in
science classes, we must provide more guidance about the nature of scientific explana-
tions and more insight into how teachers and students can generate and evaluate scientific
explanations. This is not merely an exercise in terminological precision. While clarity of
language and meaning is instrumental for communication and development of new ideas,
more important to the field of science education is a shared vision of what kinds of learning
are possible in classrooms, how these articulate with authentic disciplinary practices, and
how these can be scaffolded in principled ways for students.
Philosophers of science have long examined the structure and role of explanation in
the sciences beginning with the seminal work of Hempel and Oppenheim (1948) and
spanning the subsequent contributions of Salmon (1978, 1989), Van Fraassen (1980),
Friedman (1974), Kitcher (1989), and others. This body of work offers lenses for analyzing
how scientific explanations are constructed in science and how science educators might
design learning environments to foster scientific explanation with students in secondary
classrooms. We use the major models of scientific explanation from the philosophy of
science literature to engage with four questions:
1. How is scientific explanation defined within philosophy of science?

2. How has research in science education classrooms framed and supported scientific
explanation?
3. How have science education reform documents framed scientific explanation?
Science Education
SCIENTIFIC EXPLANATIONS 641
4. How has our research project supported teachers’ and students’ work with scientific
explanations over the past four years?
FIVE MODELS OF SCIENTIFIC EXPLANATION FROM PHILOSOPHY

OF SCIENCE
Philosophers of science have spent decades analyzing both the process of constructing
explanations and the merits of existing scientific explanations themselves. There is no
unitary theory of explanation in philosophy of science, but there are some areas of general
agreement. Many philosophers of science broadly conceptualize scientific explanations as
attempts to move beyond descriptions of observable natural phenomena into theoretical
accounts of how phenomena unfold the way they do (Achinstein, 1983; Kitcher & Salmon,
1989; Nagel, 1979; Salmon, 1978, 1989).
Scientific explanations for natural phenomena often involve unseen entities such as atoms
or forces, underlying processes such as genetic drift or oxidation, statistical or probabilistic
patterns, or broad scientific theories to account for natural phenomena (Friedman, 1974;
Kitcher, 1989, 1997; Salmon, 1978, 1989). For example, a description of condensation
appearing on the outside of a cold glass of water differs from an explanation for condensation
in that the description emphasizes observable features of the phenomenon such as the cooler
temperature of the water in the glass and the presence of droplets on the outside of the glass.
In contrast, an explanation for condensation emphasizes unobservable processes such as
molecular motion and energy, employs key scientific ideas and theories, and often seeks
underlying causes for a commonly observed phenomenon.
Both scientists and science students benefit from seeking scientific explanations. Salmon
(1978) highlights the advantages for scientific thought of making a move from descriptive
to explanatory accounts in an address to the American Philosophical Association:
It provides knowledge of the mechanisms of production and propagation of structure in
the world. That goes some distance beyond mere recognition of regularities, and of the
possibility of subsuming particular phenomena thereunder. It is my view that knowledge
of the mechanisms of production and propagation of structure in the world yields scientific
understanding, and that this is what we seek when we pose explanation-seeking why
questions. The answers are well worth having. That is why we ask, not only “What?” but
“Why?” (p. 701)
For science teachers working with middle school and high school students, a focus on
scientific explanations affords an opportunity to teach for understanding, but this kind of
focus is challenging for the majority of science teachers (Driver et al., 1994; Gamoran
et al., 2003; Treagust, Chittleborough, & Mamiala, 2003). We have found that familiarity
with philosophical models for scientific explanation helps science teachers develop deeper
understanding of science content informing their decision making about what to teach and
how to teach in science classes (Thompson, Braaten, & Windschitl, 2009b; Thompson,
Windschitl, & Braaten, 2008). From a philosophical perspective, there are many ways
of conceptualizing scientific explanations, all of which can be relevant for research and
practice in science education. After a brief review of each of the major models of scientific
explanation from philosophy of science, we will describe how each can provide important
insights for science educators.
The “Covering Law” Model of Scientific Explanation

Covering Law Explanations in Philosophy. The original model for scientific explanation
was the deductive–nomological (D–N) or “Covering Law” model of explanation put forth by
Science Education
Hempel and Oppenheim (1948). These philosophers saw science as a discipline illuminating
regularities of the natural world that could then be expressed in statements as “natural laws.”
When seeking an explanation for an event in the natural world, scientists look to natural
laws that can account for particular events as logical, expected outcomes based on well-
established patterns. Once a natural law is expressed in a field of science, it can be used to
explain events “covered” by that law. For example, students in a chemistry class frequently
use laws such as Boyle’s law to explain particular relationships between the volume of a
gas and the pressure necessary to maintain that volume. Laws often provide mathematical
means of representing persistent patterns observed in nature, and students use these laws to
perform calculations as part of their explanation showing how specific observable events
are logical outcomes of well-known patterns.
The Covering Law model for scientific explanations has been subjected to a number
of critiques by philosophers who find fault with the symmetry of the explanations, with
the inability to account for unlikely events, and with difficulty explaining events that are
not covered by a natural law. Covering Law explanations are often confusing because
causes and effects of phenomena can seem interchangeable—a problem often referred to as
symmetry within the explanation. Salmon (1989) provides the example of observing a rapid
drop in the reading on a barometer and then inferring that a storm is approaching because
whenever the barometric pressure drops a storm follows. It does not make sense to say that
the change in the reading on the barometer explains the occurrence of the storm, but the
statement fits the Covering Law model of explanation because an event is explained as the
logical outcome of an observed regularity. Salmon points out that our intuition in this case
is to seek a common cause for both the storm and the barometer reading to provide clear,
asymmetrical explanations for the developing storm and the falling barometer reading.
Philosophers also find fault with the Covering Law model because it does not accom-
modate explanations of unlikely events or of events that are not governed by natural laws.
Cartwright (1997) argues that reliance on laws for constructing scientific explanations is
problematic primarily because very few “covering laws” exist, especially outside of physics.
Instead of relying on laws, many of the scientific explanations that are generally accepted
by scientists employ generalizations that resemble laws. Cartwright maintains that this use
of “false laws” should not detract from the quality of the scientific explanation. Instead,
these statements tend to “express our explanatory commitments” helping to define what
does and does not count as a possible explanation (p. 163). For example, it is not uncom-
mon for teachers and students in physical science classrooms to explain the solubility of
polar solutes in polar solvents and nonpolar solutes in nonpolar solvents by appealing to
a “law-like” statement that “like dissolves like.” While this “law-like” statement may be
helpful for reminding students about some rules of thumb in chemistry, it does not fully
explain exactly what is happening at an atomic or molecular level when certain solutes
dissolve in certain solvents. However, for science educators, the “like dissolves like” rule
may signal certain explanatory commitments that are worth pursuing in greater depth such
as pressing students to further unpack the important role of polarity in solubility. Because of
the numerous critiques of the Covering Law model of explanation, philosophers of science,
including Hempel himself, have considered additional models for scientific explanations
that might satisfy some of these objections and provide more satisfactory accounts for
natural phenomena (see Table 1 for a summary).
Covering Law Explanations in Science Classrooms. Critiques from Salmon and

Cartwright are important not only from a philosophical standpoint but also from a peda-
gogical standpoint because they hint at the shortcomings of “Covering Law” explanations
in supporting student reasoning. Chemistry students could rely on the gas laws to “explain”
Science Education
TABLE 1
Five Philosophical Models of Scientific Explanation Relevant for Science
Education
Models of
Explanation Attributes of an Explanation Relevance to Science Education
Covering Law • Deductive arguments • STUDENTS: Students’ first
explaining events as natural, attempts at explanations often
logical results of regularities follow this form explaining a
expressed by laws. specific event by citing a law or
• Merits depend on logical using a law-like statement.
coherence of the argument • TEACHERS: Teachers can listen
showing an event to be the for—and then extend—students’
expected result of a natural attempts at explanation.
law (see Hempel & • ISSUES: Fosters algorithmic
Oppenheim, 1948). reasoning but may not develop
students’ conceptual reasoning
or theory-building abilities.
Statistical- • Induction from a trend or • STUDENTS: Engages students in
Probabilistic pattern in data may or may not important data analysis practices
seek underlying causes for especially in fields relying on
events. large data sets like population
• Merits of explanation depend biology and earth science.
on degree of coherence • TEACHERS: Teachers can
between explanation and data engage students in data
(see Hempel, 1965; Salmon, interpretation and scaffold
1989). students’ attempts at making
inferences from data.
• ISSUES: Focuses on data and
data representations but can
divert attention away from
phenomena and mask causes of
events.
Causal • Induction from patterns in data, • STUDENTS: Capitalizes on
but explicitly seek underlying students’ curiosity and engages
causes for events. students in theorizing about
• Merits depend on coherence unobservable causes for
with data and on degree of observable phenomena.
confidence in establishing • TEACHERS: Teachers can
causation (see Salmon 1978, engage students in theorizing
1989). and model building.
• ISSUES: Involves an inherent
challenge of establishing
causation. Tendency toward
developing only simple, linear
cause–effect relationships
instead of causal webs and
models.
(Continued)
Science Education
TABLE 1
Continued
Models of
Explanation Attributes of an Explanation Relevance to Science Education
Pragmatic • Relies on shared agreement • STUDENTS: Encourages explicit
about the “contrast class” communication between teachers
inherent in the why-question: and students about locally
Why is this (and not that) the constructed norms and meanings
case? Attributes negotiated for explanations affording
and deemed acceptable by sense-making opportunities.
participants in conversation. • TEACHERS: Teachers can listen
• Varied assumptions about to students’ attempts and better
contrast class may cause understand student thinking
disagreement about what before posing follow-up
makes a satisfactory questions.
explanation (see van • ISSUES: Is not a stand-alone
Fraassen, 1980). model of explanation but offers
an important analytical tool for
teachers listening to students.
Unification • Explanations for singular • STUDENTS: Encourages students
events are unified into to focus on “big ideas” in science
generalizations through use of by emphasizing major scientific
major theories in science (i.e., theories and explanations.
kinetic molecular theory). • TEACHERS: Affords a way of
• Merits depend on degree to evaluating explanatory power of
which an idea connects student-generated scientific
otherwise disconnected explanations.
phenomena and coheres with • ISSUES: Not helpful for creating
other accepted explanations in explanations of single
the ‘explanatory store’ (see events—more useful in concert
Friedman, 1974; Kitcher, with other models.
1997).
many aspects of the behavior of gases without reasoning through underlying theoretical
ideas about the particulate nature of matter or the ability of tiny particles to exert forces
during collisions. Driver Leach, Millar, and Scott (1996) argue that “Covering Law” expla-
nations foster algorithmic reasoning, but they do not foster sophisticated reasoning or deep
conceptual understanding because students do not need to engage in reasoning beyond the
use of laws.
Driver et al. (1996) point out that many everyday ways of explaining events take the
form of “Covering Laws.” For example, the authors envision a child asking a parent why
the frying pan has a plastic handle, rather than a metal one. The parent responds that
plastic does not conduct heat, but metal does; therefore, plastic handles do not get hot
when the pan is on the stove. The parent’s explanation involves a statement asserted as
a fact in a “law-like” form similar to those suggested by Cartwright. There are certainly
more details that could be included in the explanation, but the parent’s response essentially
takes the form of a Covering Law explanation with “plastic does not conduct heat” being
used as a sort of law. Driver and her colleagues encourage teachers and students to explore
additional models of explanation to foster deeper understanding in science. However,
the Covering Law model of scientific explanation does satisfy some of our intuitions
Science Education
about explanations by using logical argument construction based on persistent patterns

observed in the natural world. In science classrooms, students often begin their attempts
at providing scientific explanations by creating statements that resemble “Covering Law”
explanations.
In our secondary science classrooms, it is common to hear students making statements
appealing to a law-like regularity as a way of accounting for something observed in nature.
For example, we recently sat down with two students who were just beginning to examine
phase changes in water. They had recently read about the behavior of molecules as ice
melts and as water is boiled. The students were looking closely at a beaker of water sitting
on a hot plate; water vapor was just beginning to rise from the surface of the water when
we asked, “So why do you think that water vapor is starting to rise from the beaker of
water?” One student replied matter of factly, “Because it’s on the hot plate heating up, and
that’s what happens right before it boils.” When we think about this student’s attempt at
explaining vaporization, we can see that she is using her knowledge of prior patterns when
she remarks “because that’s what happens when. . . ” and she is explaining the vaporization
using this pattern as a kind of “law.”
By recognizing the law-like quality of students’ initial attempts at explanation, we are
given a starting place for pressing students to move beyond chalking up a phenomenon
to a law or a generalization when it is often possible for them to dig deeper and think in
terms of underlying causes or in terms of powerful theories. Rather than simply criticizing
students for failing to fully explain a phenomenon, we can recognize their initial attempts at
explanation as being similar to Covering Law explanations and then give explicit types of
feedback. We can communicate with students about how their initial attempt at explanation
lacks certain attributes such as a causal relationship or an overarching scientific theory in
an effort to press them to construct deeper scientific explanations. Alternatively, we may
choose to accept students’ Covering Law explanations in certain circumstances, such as
during a physics unit focused on mechanics, where law-like statements are the disciplinary
norm for sufficient scientific explanations.
Statistical Model of Scientific Explanation

Statistical/Probabilistic Explanations in Philosophy. A second type of scientific ex-
planation, the statistical or probabilistic model, is sometimes seen as a subset of other
models for explanation. Some philosophers of science argue that explanations employing
statistical and probabilistic reasoning are among the most important and complex in science
(Hempel, 1965; Salmon, 1989). Phenomena such as the decay rate of radioactive isotopes,
the likelihood of inheriting a particular gene combination, or the chance of sea levels ris-
ing in response to global warming are often understood by using mathematical reasoning
rather than appealing to laws or generalizations. Mathematical constructs like rates and
probabilities may not follow universal generalizations of the sort initially imagined by
Hempel and Oppenheim (1948). In some cases, such as phenomena examined in theoret-
ical physics, the phenomena can appear random rather than regular. Statistical models of
scientific explanation are used in these cases to provide an account for phenomena that are
not “covered” by a law. For example, when it is observed that many residents of a small
town are developing an unlikely form of cancer, we cannot explain the higher incidence of
cancer using a law-like statement because the appearance of the cancer was unexpected.
Instead, we seek other explanations for the increased rate of cancer, such as exposure to a
particular carcinogen present in the area. By connecting an increased incidence of cancer
to an increased exposure to a carcinogen, we create a statistical explanation built from
a correlation in data. However, there is also an underlying biochemical process at work
Science Education
when a carcinogen causes cancer. Examining data to arrive at a statistical or probabilistic

explanation only partially assists us in explaining how certain phenomena work. This small,
but consequential, detail reveals a concern about relying on statistical models of scientific
explanation.
Salmon (1989) raised objections about statistical scientific explanations because he was
concerned that emphases on statistics might gloss over the actual underlying explanation
for events. He illustrated his objections with the classic example where a doctor notices that
a patient recovered from a cold after a week of taking vitamin C supplements. The doctor
then conducts a study and finds that all patients recovered from their colds after a week
of vitamin C treatments, which seems to imply that vitamin C is the cure for the common
cold. Unfortunately, by focusing on data about vitamin C treatments we have overlooked
the important detail that patients will usually recover from colds within a week regardless
of the vitamin C because colds are caused by self-limiting viruses that healthy immune
systems can usually handle. Salmon argues that in this case, probabilistic reasoning might
actually impede efforts to understand deeper explanations, but he allows that there are many
areas of science where statistical and probabilistic scientific explanations are critical for
understanding phenomena.
Statistical/Probabilistic Explanations in Science Classrooms. In school science con-

texts where students are examining phenomena by using large data sets, statistical scientific
explanations may play an important role, a point emphasized by Duschl (1990, 2000).
For example, in a science class focusing on population genetics, probabilistic data can
be used to make sense of variations in gene pools across time and geography, as well
as predict future trends. From our own recent study of novice teachers’ development of
practice (Thompson et al., 2009a 2010; Windschitl, Thompson, & Braaten, 2008a, 2011)
one teacher named Sarah pursued such an explanation during a genetics unit in her high
school biology class. Students were engaged in examining two data sets: (1) a map of Africa
showing allele frequencies of sickle cell anemia and (2) a second map of Africa showing
epidemiology data about malaria. Sarah engaged students in an investigation into the rela-
tionship between these two diseases. At the end of the unit of study, students were asked to
use probabilistic reasoning with the data sets and to use their understanding of inheritance
patterns and evolutionary theory to explain the continued existence of a harmful gene in
a population. Sarah’s example illustrates that in many cases where statistical explanations
might be warranted, teachers and students could also be focusing on the complex causes
for events, the mediating factors that complicate natural systems, and the central theories in
science that provide the underpinnings for our current understanding of phenomena. This
strongly suggests that teachers will need to be knowledgeable and flexible in their ability
to juggle multiple models of scientific explanation simultaneously requiring a special kind
of pedagogical content knowledge that Treagust and Harrison (2000) term “explanatory
flexibility” (p. 1161).
Causal Model of Scientific Explanation

Causal Explanations in Philosophy. Both the Covering Law model of scientific expla-
nation and statistical models of scientific explanation hint at, but do not fully satisfy, a strong
intuition about scientific explanation—our desire to understand the causes for events in na-
ture. Salmon (1978) suggests that scientific explanations seem most compelling when they
offer understanding of the causes underlying natural phenomena. While the other models
of scientific explanation may include causal relationships, they do not explicitly emphasize
Science Education
causation as a key attribute of explanatory power. Salmon argues that explanatory power
is enhanced when causal scientific explanations involve ideas from scientific theories to
account for phenomena, such as using gravitational theory to explain the regularity of tides
or using kinetic molecular theory to explain the behavior of gases. Both tides and behaviors
of gases can be readily described by laws using the Covering Law model of explanation, but
the desire for a causal account is better satisfied when theories and underlying mechanisms
are included in the explanation. The causal model of scientific explanation has advantages
in that it often resolves some of the puzzles of the Covering Law model, including the
symmetry problem and the problem with explaining unexpected events; however, causal
scientific explanations rely on our ability to establish causation, which is a puzzle that
has plagued philosophers for well over a century and continues to stir debate (Salmon,
1989). The principal challenge of a causal model of scientific explanation lies in locating
and substantiating underlying causes for phenomena, which requires a degree of infer-
ence to establish connections between causes and effects. Despite these difficulties, the
causal model of scientific explanation remains the preferred model for many disciplines of
science.
Causal Explanations in Science Classrooms. Many of the phenomena examined in

school science classes have well-established causal explanations that form the conceptual
bases of many disciplines in science. In microbiology, for example, scientists are almost
always seeking causes for phenomena such as the infectious agent responsible for disease
symptoms, the epidemiology of a particular outbreak of a disease, or the chemical pro-
file of soil samples during bioremediation. Without causal scientific explanations, neither
the student nor the microbiologist would really understand phenomena such as disease,
epidemics, or bioremediation. However, just as philosophers of science find causation to
be conceptually and philosophically challenging, science teachers face challenges when
pressing for causal explanations in science classes.
Perkins and Grotzer (2005) note that both teachers and students are able to propose
simple, linear cause–effect models to explain phenomena, but they often have a difficult
time proposing more complex storylines that include interrelated causes and mediating
factors that more accurately represent the kinds of explanatory models employed in bio-
logical sciences. However, when students are able to craft complex, causal explanations
for biological phenomena, their understanding of the phenomenon and of the underlying
biological ideas are enhanced (Grotzer & Basca, 2003). In our research, we have found
that by providing novice science teachers with some background knowledge about causal
scientific explanations and then pressing them to wrestle with these explanations dur-
ing their science classes, we can help teachers identify the major scientific ideas worthy
of in-depth inquiry for their students (Thompson et al., 2009a, 2010; Windschitl et al.,
2008a).
One of the teachers in our study, Simon, often considers the full causal explanation for
phenomena when he plans units of instruction designed to help Grade 9 students understand
major science ideas deeply (Thompson et al., 2010; Windschitl et al., 2011). For example,
during Simon’s second year of teaching he decided to modify the “Sound and Waves” unit
of instruction. The science textbook suggested covering every kind of wave and focusing
students on calculating wavelengths, frequencies, and amplitudes, but Simon felt that none
of these isolated ideas helped students explain the differences in sounds that they could
hear and feel when attending concerts. As a result, he engaged students in investigations
into the movement of sound energy through the medium of air and examinations of the
intensity and pitch of sounds using music software on their computers. Students developed
Science Education
rich explanatory models to account for the sounds that they heard coming from musical
instruments and were able to expand those models to think about other phenomena such
as echoes and movement of sound through another medium such as water. A focus on
causal explanations has helped Simon, and many of our other beginning teachers, make an
important move toward organizing units of instruction around “Big Ideas” in science and
focusing students’ science investigations on explanation and theory-building in addition to
observation and description (Windschitl et al., 2010).
Pragmatics of Explanation
Pragmatics of Explanation in Philosophy. Philosophers taking a pragmatic view of
explanation do not suggest a specific model for scientific explanation. For those taking
a pragmatic view of explanation, the context surrounding the request for an explanation
determines whether or not a response “counts” as a satisfactory explanation. In other words,
the adequacy of an explanation stems not just from features of the explanation, but from
the question initiating the explanation as well as the context surrounding the question
(Achinstein, 1983). By looking at the whole context in which an explanation is constructed,
we can understand how groups of people construct explanations together in social settings
similar to those examined in studies of practicing scientists (Knorr-Cetina, 1999; Latour &
Woolgar, 1986; Traweek, 1988).
This fourth view of explanation is well articulated by Bas Van Fraassen (1980), who
critiques other attempts at developing a singular, unified theory for scientific explanation
because these attempts fail to place explanations in context. He argues that when context
is taken into account, the standards for judging the merits of explanations are not solely
about the alignment between theory and facts but are also negotiated between participants
who are constructing the explanation in a particular context. For example, multiple people
could discuss a recent car accident at a three-way intersection seeking an explanation
for the accident. Owners of businesses on the corners of the intersection might explain
the crash by citing the continued hazards posed by the awkward intersection design. The
city manager might explain the accident by citing the need for a traffic signal to regulate
drivers at the busy intersection. The driver of one of the cars might explain the accident
by noting his difficulty with blind spots in his car. Each explanation, for philosophers who
take the pragmatic view of explanation, is acceptable so long as the local participants in the
conversation determine that the explanation satisfies their request for an explanation.
Van Fraassen (1980) argues that the key factors in determining the adequacy of ex-
planations are the “contrast class” and the “relevance relations” that are emphasized by
participants in a discussion. The local business owners emphasized that this particular acci-
dent is simply the logical outcome of the many near-accidents that happen at the awkward
corner. The contrast class, in the case of these local residents, is the set of near-accidents and
the detail that is most relevant is the unusual convergence of three streets at the intersection.
However important these details are to the group of local business owners, they are not
important for the driver who is more focused on the problems caused by blind spots in his
car, which could have caused him to crash at this or any other intersection. Differences in
contrast class and relevance relations cause people to propose different explanations for the
same event that are equally acceptable and valid, but that can sound completely irrelevant
to different participants in a conversation.
Pragmatics of Explanation in Science Classrooms. Understanding Van Fraassen’s ar-

guments and the pragmatics of explanation has helped our support of science teachers and
Science Education
students in important ways. By examining the multiple interpretations of both the request
for explanation and the responses provided by our secondary science students, we can see
the importance of developing shared understanding about scientific explanations. For exam-
ple, in a sixth-grade physical science class, the first author worked with students to explain
the movement of water through the hydrologic cycle. During a class period when students
were drawing upon recent experiments with phase change and a reading about the molecular
motion and energy transfer, the first author asked one student, “Why do water molecules
move from lakes, into the atmosphere, and then into clouds?” The student responded that
water moves this way because animals and plants need water to survive. Although the
teacher had frequently emphasized attributes of causal explanation and provided scaffold-
ing prompting students to include ideas about unobservable, underlying mechanisms, the
“why” question itself lacked specificity and failed to communicate the kinds of ideas that
the teacher “counts” in an explanation. It makes sense, then, that the student’s response
does not include ideas about molecules, energy, or phase change and focuses instead on
how living things require water. Thinking about explanation from a pragmatic perspective
has helped us realize how important it is to reframe our questions to clarify the contrast
class and make explicit the kinds of details relevant for a good explanation. In many cases,
including this one, a follow-up conversation was necessary to co-construct the kinds of ideas
that “count” in a good explanation about phase change. Developing scientific explanations
as well as negotiating what “counts” as a good scientific explanation within the context of
a science classroom has afforded an opportunity for shared knowledge construction that is
not otherwise present.
Explanatory Unification View of Scientific Explanation

Explanatory Unification in Philosophy. A final model for scientific explanation is the
explanatory unification view advanced first by Friedman (1974) and later by Kitcher (1989,
1997). This perspective builds upon prior conceptualizations of scientific explanation and
contends that explanatory power is increased when explanations can unify seemingly dis-
connected phenomena into a coherent relationship, providing “global rather than local”
understanding of phenomena (Friedman, 1974, p. 18). Powerful theories in science, such
as the kinetic molecular theory, derive their explanatory power from their ability to pro-
vide explanations for different phenomena across many contexts. For example, instead
of relying on one explanation to account for the way that heated air inflates a hot air
balloon and then employing a different explanation to account for the condensation of
water vapor on the outside of a cold glass, we can use the kinetic molecular theory to
account for both of these seemingly unrelated phenomena. Instead of providing an en-
tirely new model for scientific explanation, explanatory unification provides a way of
evaluating the explanatory power of theories and conceptualizing how powerful scientific
explanations deepen scientific understanding in ways that descriptions and descriptive laws
cannot.
Friedman (1974) argues that powerful scientific explanations should promote deeper
understanding of the natural world. He maintains that if scientific explanations are simply
logical arguments about the natural world, as suggested by proponents of the Covering Law
model, then providing a scientific explanation does nothing more than replace “one brute
fact with another” (p. 14). Instead, Friedman suggests that powerful scientific theories,
like the kinetic molecular theory, afford the ability to apply explanations for phenomena
across a range of observations and help us to unify phenomena and their explanations.
It is this unification, according to Friedman, that fosters understanding because “a world
Science Education
with fewer independent phenomena is, other things equal, more comprehensible than one
with more” (p. 15).1 The link between explanatory unification and the ability to better
understand the natural world suggests an important role for scientific explanation in science
education.
Explanatory Unification in Science Classrooms. The explanatory unification view of

scientific explanation is important for science education for three reasons: (1) the emphasis
is on sense making and understanding, (2) the focus is on big science ideas, and (3)
there are clear ways to evaluate explanatory power. School science is often described as
covering a range of topics that are a “mile wide and an inch deep” (Duschl & Grandy, 2008;
NRC, 2007). The explanatory unification view demands that scientific explanations focus
on overarching scientific theories that apply globally to phenomena not just locally. This
view of scientific explanation offers a way for teachers to begin to pare down sprawling
science curricula and focus intentionally on the most important ideas in each scientific field.
The explanatory unification view of scientific explanations offers a means of evaluating
the explanatory power of ideas in science—a valued theory-building practice emphasized
in many science education reform documents (AAAS, 1993; NRC, 1996, 2007). Kitcher
(1997) proposes the notion of the “explanatory store,” a set of explanations that already exist
and are commonly held in scientific thought (p. 170). He suggests that newly constructed
scientific explanations often draw from or are judged against ideas in this store. The notion
of the “explanatory store” could further assist teachers in deepening the curricula to include
the most central ideas in science and would offer a way of guiding students to evaluate
explanatory power.
Summary of Models of Explanation

These five major models of scientific explanation from the philosophy of science are
all relevant for science educators, not because of some priority given by philosophers, but
because each is a legitimate practice of scientists and children attempting to craft expla-
nations. Conceptual clarity about scientific explanations helps science educators envision
what “counts” as a big idea in science lessons and units of instruction. A framework in-
formed by philosophy of science also enables science educators to make sense of the partial
attempts at explanation provided by students, which, in turn, allows teachers to make further
decisions about how to proceed with instruction.
Most science educators in the United States, however, have not been immersed in phi-
losophy of science ideas. Instead, many science educators look to literature in science
education for guidance about what to teach and how to teach. The science education reform
literature from the past two decades provides normative visions for what science educa-
tion should be, whereas the literature from research in science classrooms illuminates how
specific pedagogical practices, tools, and curricular innovations impact students’ learning
experiences. We now turn to these two bodies of literature to see how the field of science
education characterizes scientific explanations, how science educators are guided to recog-
nize the power of certain kinds of explanations, and how science educators are helped to
1
Salmon (1978) cautions that the concept of “understanding” belongs more to the domain of psychology
than to philosophy, a distinction also offered by Hempel and acknowledged by Friedman, but Salmon
concedes that powerful scientific explanations do satisfy many of our intuitions about understanding the
natural world.
Science Education
analyze and evaluate the merits of alternate explanations—in particular those provided by
their students.
CONCEPTUALIZATIONS OF SCIENTIFIC EXPLANATION

IN SCIENCE EDUCATION
Despite a strong emphasis on the role of scientific explanation in education, there is
a lack of clarity about the concept of scientific explanation within the science education
literature. Because it is the science education literature—not the philosophy of science
literature—that ultimately informs science educators, conceptual clarity is essential for
guiding instructional practice. In particular, the science education community needs more
conversations about three overarching questions: (1) What constitutes a “good” scientific
explanation in a science classroom?, (2) What makes an explanation explanatory rather
than descriptive?, and (3) How might we evaluate the merits of alternate explanations
offered by students in classrooms? Using philosophy of science as a lens, we examined
recent research and reform efforts focused on developing scientific explanations in class-
rooms to better understand how the science education community conceptualizes scientific
explanation.
Scientific Explanations in Science Education Research

Researchers in science education have conducted numerous studies designed to engage
teachers and students in constructing, analyzing, and evaluating scientific explanations.
However, many of these intervention studies do not specify their conceptualizations of
scientific explanations in ways that can provide further guidance for other researchers
and science education practitioners. Some studies present the construction of a scientific
explanation as straightforward and unproblematic, whereas others focus on only one aspect
of explanation. In some cases, scientific explanation is combined with argumentation. In
the following sections, we describe the ways in which explanation is portrayed in science
education research (see Table 2). Although the nature of explanation is typically consistent
within various programs of research, it differs significantly across the field resulting in
conceptual ambiguity.
Explanation as Explication. In the science education literature, it is common to see

“explanation” used in the sense of providing clarification for the meaning of a term
or explication of reasoning about a problem. In science classrooms, students are fre-
quently asked to “explain their reasoning” while solving a problem, to “explain the
meaning” of a technical term, or to “explain the results” of an experiment. Providing
an explanation—or an explication—is in many ways an authentic communicative prac-
tice in the daily work of scientists who clarify ideas and findings to various audiences
(Knorr-Cetina, 1999; Latour & Woolgar, 1986; Traweek, 1988). While it is important to
engage in this kind of clarification to communicate, the practice of constructing scien-
tific explanations that account for natural phenomena involves more than explications of
meaning.
Ogborn, Kress, Martins, and McGillicuddy (1996) provide a thorough analysis of the
ways that teachers and students engage in explication in the science classroom. Teachers
or students might, for example, help another student to understand the meaning of un-
familiar vocabulary by explaining the term in a different way. The authors also point to
Science Education
TABLE 2
652
Common Uses of “Explanation” in Science Education
Uses of
“Explanation” in Attributes of This Kind of Examples in Science Ways This Use of “Explanation”
Science Education Explanation Classrooms Can be Problematic
Explanation as • Definitions of terminology requested In classroom discourse: • The object of the explanation—what is
explication by teacher or students, prompted by • Can you explain what you mean being explained—is unclear.
instructional materials. when you say that the sugar Explications are focused on
• Metacognition about reasoning/ “dissolved”? terminology, meaning, and reasoning
problem-solving strategies • Can you explain how you figured out while scientific explanations are
requested by teacher or students, the amount of force needed to lift focused on natural phenomena.
prompted by instructional materials. that load with the pulley system? • Teachers and students engage in
In instructional materials: explication frequently, and may think
BRAATEN AND WINDSCHITL
• Explain what biological evolution that they are working on scientific

means. Include the following: explanations when they are not.
adaptation, population, natural
selection, variation, species,
mutation.
• Explain how your procedure
illustrates the law of conservation of
mass.
Explanation as • Emphasis on cause-effect In classroom discourse: • Oversimplifies the nature of causation
simple causation relationships accounting for an • What’s making the sugar dissolve so into simple cause-effect relationships
observable event. fast? instead of complex webs of causation.
• Why does the cart roll faster on the • Implies that causal explanation is the
smooth ramp than on the carpeted single, accepted model of scientific
ramp? explanation when there are actually
many acceptable models for scientific
explanations.
(Continued)
Science Education
TABLE 2
Continued
Science Education
Uses of
“Explanation” in Attributes of This Kind of Examples in Science Ways This Use of “Explanation”
Science Education Explanation Classrooms Can be Problematic
In instructional materials:
• What effect does “capturing” a
particular color dot have on the
numbers of that color in the following
generations?
• What effect did the salt have on the
boiling point of water?
Explanation as • Emphasis is on argument In classroom discourse: • Emphasis on scaffolding students’ use
justification construction—often taking the form • How do you know that this gas is of evidence to justify claims overlooks
of “claim-evidence-reasoning.” carbon dioxide? possible explanatory power of claims.
• Requests for “explanations” are • What makes you so sure that no • Claims prompted by these efforts are
requests for evidence and reasons other forces are acting on the cart? often assertions or descriptions rather
for belief in a claim. In instructional materials: than explanatory (how/why) claims.
• What evidence was there that • Successful scaffolding for explanatory
chemical reactions were occurring? claims is downplayed and therefore
• Work with the members of your team invisible to other science educators.
to determine which explanation best
accounts for the data presented.
SCIENTIFIC EXPLANATIONS
653
the importance of helping students “see” theoretical entities and actions in the same way
that scientists see them through the use of explication. In a chemistry or physics class, for
example, teachers rely on explication to help students envision subatomic particles, waves,
energy, or forces acting across multiple unobservable contexts. In this sense, Treagust and
Harrison (2000) bridge the practices of creating scientific explanations and providing ex-
plications with their concept of a science teaching explanation. Science teachers often use a
variety of discursive strategies including metaphors, analogies, and vignettes to communi-
cate well-established scientific explanations to students. Treagust and Harrison recommend
that teachers become well versed in a number of “explanatory frameworks” to provide
better science teaching explanations in classrooms (p. 1158). These episodes of clarifica-
tion are not in themselves examples of scientific explanations, but they may be integral to
understanding the discourse moves required in classrooms where teachers and students are
working to understand scientific explanations. Ogborn et al. and Treagust and Harrison are
focusing primarily on instances when teachers explicate an idea to their students; however,
it is also important to consider how students themselves might talk through their own ideas
about science.
For students to make sense of science ideas like particles or energy, and engage in
construction of explanations, we must create opportunities for students to talk about
science—to explicate aloud (Herrenkohl, 2006; Lemke, 1990; Mortimer & Scott, 2003).
In such cases, explication may be a necessary step when helping students construct sci-
entific explanations for phenomena, but it is not synonymous with scientific explanation.
Much of the focus on scientific explanation in schools tends to portray “explanations”
as knowledge produced by scientists and then explicated to students by teachers through
skillful use of representations, stories, discussions, and analogies (Millar & Osbourne,
1998; Treagust et al., 2003). We would like to extend Treagust and Harrison’s (2000)
notion of building explanatory frameworks with teachers to include building explanatory
frameworks with students as well. By working together to establish what “counts” as a
good explanation depending on contextual factors such as disciplinary norms and science
ideas available to students at certain times of year or in certain courses, teachers and stu-
dents could develop shared understanding and further the knowledge construction in their
classrooms.
The term “explanation” is also used to connote the communication of reasoning in an
effort to make thinking visible or audible in science classrooms. Coleman (1998) examined
students’ explanations in science as part of a study on student learning in problem-based
science classes; however, the explanations that were examined were not scientific explana-
tions. Rather, students were prompted to explain their reasons for thinking that a particular
answer to a question was justified, to defend why they believed their answers to be correct,
or to reflect on changes in their thinking over time. During conversations with students, for
example, Coleman describes how a prompt such as “Ok, explain why you believe that your
answer is correct or wrong?” pushes students to communicate their reasoning (p. 406). Sim-
ilarly, a prompt such as “Can you compare how you used to think about this with how you
think about it now?” helps to further conversations with students (p. 407). Berthold, Eysink,
and Renkl (2009) and Wittwer and Renkl (2008) found that having students construct these
kinds of explanations of reasoning on their own or with a peer offers an opportunity for
sense making that does not seem to happen when students simply listen to an instructional
explanation from a teacher or read one in a text. Creating opportunities for students to
verbalize their thinking with peers and with teachers is critical to helping students make
sense of science ideas and may be key components to helping students engage in productive
classroom discourse, but this type of classroom task does not appear to call for scientific
explanation itself.
Science Education
The confusion created by the use of the term “explanation” as a synonym for clarification
or for communication of reasoning is not merely one of semantics. The problem is that
the object of the intellectual work—what is being explained—is different for each sort of
explanation. Scientific explanations seek to explain events in the natural world. However,
when teachers press students to explicate the meaning of words or to communicate their
thinking about a problem or their procedures for arriving at solutions to problems, the object
of the “explanation” is no longer natural phenomena. In addition, because teachers request
“explanations” often in classrooms, they may believe that they are pressing for scientific
explanations when, in fact, there may be nothing explanatory about the intellectual work
at all. To provide the conceptual clarity necessary for supporting science educators, it is
important that we tease apart the important everyday discursive activities of clarifying terms
and communicating reasoning about ideas from the specific science discourse practice of
crafting scientific explanations.
Explanation as Causation. Some scholars in science education take a perspective that

to explain a natural phenomenon we must establish a causal account using underlying
mechanistic properties of the natural world to explain observable phenomena (cf Windschitl,
2008; Hammer, Russ, Mikeska, & Scherr, 2008; Russ, Scherr, Hammer, & Mikeska, 2008).
Hammer et al. make the strong claim that causal, mechanistic accounts of phenomena are
the principle kinds of explanations in science. For many disciplines of science, developing
causal explanations are indeed central practices for making sense of phenomena, and many
of the big ideas addressed in school science have well-known causal explanations. There
are certainly branches of science in which causal explanations are not the coin of the realm.
Fields such as population genetics and quantum physics utilize statistical and probabilistic
reasoning to make sense of phenomena for which there may not be any underlying cause
or regular mechanism. Other fields, such as classical physics, employ laws—statements of
observed regularities—rather than underlying causes to account for the operation of simple
machines or to describe the motion of certain objects. By pressing for causal explanations
exclusively, science educators may be misrepresenting the kinds of scientific explanations
actually employed in the sciences.
Looking to the disciplinary norms of practicing scientists can illuminate both the fine-
grained differences between explanation practices in subdisciplines of science and point to
the explanatory practices that science fields have in common. When researchers in science
education make clear these disciplinary norms, it helps science educators better understand
how to create such science learning environments. For example, Grotzer and her colleagues
(see Grotzer, 2003; Grotzer & Basca, 2003; Perkins & Grotzer, 2005) study classrooms
emphasizing complex causal models for biological phenomena. The researchers noticed
that students and teachers tended to explain ecological relationships with simple, linear
cause–effect explanations. However, they wanted students’ explanations to more closely
resemble those used by ecologists—nuanced and complex webs of explanations employing
a combination of causal and statistical/probabilistic explanations. As a result, Grotzer and
her colleagues worked to press teachers and students to combine causal and statistical
reasoning.
Palincsar, Anderson, and David (1993) also appealed to the disciplinary practice of
scientists when they chose to emphasize causal explanations employing unobservable
or theoretical entities to account for change of state, solubility, and other physical and
chemical phenomena common to school science. Like Hammer and his colleagues (2008),
the intervention of Palincsar et al. focused on crafting science instruction pressing
Science Education
students to create a particular kind of scientific explanation—a causal explanation using

unobservable underlying mechanisms—but Palincsar et al. provided insight into their way
of conceptualizing scientific explanations, thus giving guidance to other science educators.
When scholars in science education communicate the rationale behind a choice to
focus on a specific kind of scientific explanation, their work can provide both a rich,
discipline-specific conceptualization of scientific explanation and guidance for pedagogical
decision-making useful to other science educators. However, it is important to
consider how using multiple models of scientific explanation together might foster au-
thentic epistemic practices and complex reasoning in ways that total allegiance to a sin-
gular model of scientific explanation cannot. Being able to provide a causal account of
the natural world satisfies an intuition about scientific explanation that has been repeatedly
emphasized by philosophers of science; however, our point here is that establishing some
clarity of meaning for explanation does not necessarily require allegiance to a singular form
of explanation. Instead, we are advocating an explicit conversation within the field about
what makes explanations explanatory, rather than descriptive, and about how teachers and
students can engage in generating, rather than only justifying, explanations.
Explanation as Justification. Scientific explanation and argumentation are key practices

in science with a long intellectual history of analysis by philosophers of science and a recent
surge in analysis by researchers in science education (Berland & Reiser, 2009; Bricker
& Bell, 2008; NRC, 2007). Berland and Reiser point out that many science education
researchers treat explanation and argumentation as a single practice because they are so
interconnected epistemically in terms of using evidence and logic as part of a specialized
rhetorical genre. But they also suggest that both teachers and students may benefit from
educational supports that tease apart aspects of scientific explanation from argumentation. A
number of studies in science education have sought ways of supporting teachers and students
in constructing and defending scientific explanations—these interventions emphasize both
argument construction and the development of powerful explanatory ideas; however, the
reader must work to recognize the distinctions. When, for example, is one engaged in the
work of argument construction, and when is one working to develop explanatory ideas?
Osborne (2010) and Bell (2010) recently highlighted the challenge that this disentanglement
presents for science educators who are trying to make an explicit choice to focus science
learning on a few particular scientific practices—making a choice of practices often means
that some worthwhile practices lose emphasis while others receive more explicit emphasis.
Bell recommends that we take time to carefully consider a variety of scientific practices by
examining students’ everyday practices, by examining scientific practice, and by exploring
how certain practices offer opportunities for sense-making about science ideas.
Recently, a number of worthwhile interventions have addressed teachers’ and students’
work with scientific explanations and argumentation. While these interventions have made
major strides toward supporting shifts in reform-oriented science teaching, they have tended
to prioritize argumentation at the expense of explanation. McNeill and Krajcik (2008) along
with other colleagues (McNeill, Lizotte, Krajcik, & Marx, 2006; Moje, Peek-Brown, Suther-
land, Marx, & Krajcik, 2004) describe scientific explanation as synonymous with scientific
argumentation. Curricular materials for this intervention define scientific explanations as
statements consisting of “three components: claim, evidence, and reasoning” (McNeill,
2009; McNeill & Krajcik, 2008). McNeill et al. (2006) are careful to prompt readers that
“explanations often refer to how or why something happens” (p. 155). This reference to
“how or why” is subsequently lost within the intervention as it shifts emphasis toward
supporting students’ argument construction. For example, students are prompted to “write
Science Education
a sentence that states whether or not boiling is a chemical reaction” and later to “write a
sentence that states whether the mass stayed the same or not” (p. 165). Students are then
prompted to provide pieces of evidence to substantiate their claims and to write statements
illustrating their reasoning about the evidence. Similarly, Songer, Kelcey, and Gotwals
(2009) used the “claim, evidence, reasoning” format to scaffold sixth-graders’ attempts at
explaining their responses to questions such as “Is this animal an insect?” (p. 615). These
interventions provide much needed support for students’ attempts at justifying claims by
using evidence, but the students’ claims are not explanatory. How can the field of science
education capitalize on these worthwhile interventions scaffolding students’ argumentation
practices while also helping students build strong explanatory claims? How can we help
students engage in theorizing as well as justifying claims?
Sandoval and his colleagues (Sandoval, 2003; Sandoval & Millwood, 2005; Sandoval
& Reiser, 2004) report findings from their studies of an intervention designed to help stu-
dents construct evidence-based explanations of the biological diversity of finches on the
Galapagos Islands employing a tool called ExplanationConstructor. Sandoval (2003) con-
ceptualizes scientific explanations in this study as “plausible causal accounts” employing
data such as finch population sizes and observations such as beak shape to explain finch
diversity in a manner consistent with the theory of natural selection. Central to Sandoval’s
framing of scientific explanation is the requirement of “causal coherence” meaning that
students’ explanations should be internally consistent as well as consistent with generally
accepted scientific principles and theories. To reach these educational goals, the software
poses specific questions prompting students to think about causation and directs students’
attention toward salient data, observations, or aspects of scientific theories that are partic-
ularly relevant for the explanation of this phenomenon.
The researchers found that students were highly successful at constructing explanatory
claims, but, because students struggled with providing evidence and reasoning to support
those claims, the researchers’ focus remained fixed on aspects of argumentation that seemed
most problematic for these students. Sandoval (2003) focused on students’ coordination of
evidence with theory and downplayed students’ successful construction of solid explanatory
claims, missing an opportunity, we believe, to highlight some very important scaffolding
moves embedded in the tool.
Again, our critique here is not with the way that the researchers supported student
thinking through rigorous and thoughtful intervention. Instead, our critique is that this
research often features argument construction over scientific explanation as the focus of
the scholarly work. Most science educators will not use robust teaching tools; instead,
they will design instructional sequences using existing curricular materials and personal
pedagogical decision making. To provide guidance to teachers hoping to replicate the
successful scaffolding provided in the complex lesson design of ExplanationConstructor and
other similar interventions, researchers should communicate their reasoning about scientific
explanations and scaffolding used to encourage students to construct rich explanations for
natural phenomena.
From these studies, it seems a conversation is needed in the field about how particular
scaffolding supports for scientific explanations are useful for supporting the development
of students’ explanatory ideas as well as argumentation practices. Currently, the curricula,
prompts, and scaffolding in these studies direct students to use evidence and reasoning
to support assertions resulting in well-articulated statements of justified belief, but not
necessarily resulting in a scientific explanation consistent with models of explanation seen in
philosophy of science. We note here that these critiques are not of the studies themselves—
the studies are some of the most thoughtful and rigorous examples of research exploring
Science Education
the boundaries of students’ epistemic practices. It is our hope, however, that attention and
guidance for scientific argumentation and for scientific explanation can be developed by
analyzing the features of both and identifying what features of thinking and discourse need
to be scaffolded for each.
Scientific Explanations in Science Education Reform Documents

Because education reform recommendations are widely used to craft instructional ma-
terials, shape instruction, and inform assessments, it is important to examine how these
documents address scientific explanations. In the United States, two reform documents, the
Benchmarks for Science Literacy (AAAS, 1993) and the National Science Education Stan-
dards (NSES; NRC, 1996), underpin instructional materials often used in teachers’ science
classrooms while a newer review of research, Taking Science to School (TSTS; NRC, 2007)
sets the tone for current science education initiatives. We focus on U.S. reform efforts first
because that is the context in which our preservice and in-service teachers work, but we
also examine documents from the United Kingdom and European Union where scientific
explanation plays a role in on-going science education reforms.
In the Benchmarks, scientific explanations are described as ideas crucial for making sense
of scientific information and for determining the quality of scientific theories. However, the
Benchmarks do not offer any clear conceptualizations of scientific explanation, often using
the word “explanation” without providing any clarification of meaning. For example, the
Benchmarks state that “scientific investigations usually involve the collection of relevant
evidence, the use of logical reasoning, and the application of imagination in devising
hypotheses and explanations to make sense of the collected evidence” (p. 12). In addition,
the Benchmarks suggest that explanations play a critical role in the evaluation of scientific
knowledge, stating that “theories are judged by the range of observations they explain,
how well they explain observations, and how effective they are in predicting new findings”
(AAAS, 1993, p. 13). While these assertions introduce the concept of explanatory power
as a goal of theory building, they do not help to clarify what an explanation is nor do they
offer any guidance about what gives some theories more explanatory power than others.
The NSES (NRC, 1996) suggest that science educators should focus on developing,
revising, and communicating scientific explanations as part of an overall shift toward
viewing “science as argument and explanation” (p. 113). The NSES envision a growth in
the sophistication of students’ explanations as students advance through science courses,
stating that “their scientific explanations should more frequently include a rich scientific
knowledge base, evidence of logic, higher levels of analysis, greater tolerance of criticism
and uncertainty, and a clearer demonstration of the relationship between logic, evidence,
and current knowledge” (p. 117). In later sections of the NSES, the concept of scientific
explanation is differentiated from descriptions by noting that students should be able to
provide “causes for effects” and should be able to establish these cause-effect relationships
“based on evidence and logical argument” (p. 145).
Like the Benchmarks, the NSES propose that students and teachers focus on evaluating
the merits of scientific explanations as a way of understanding the growth of knowledge
of science. The rationale here is that “the scientific community accepts and uses such
explanations until displaced by better scientific ones. When such displacement occurs,
science advances” (p. 148). The criteria for evaluating alternative explanations, according
to the NSES, include logical consistency, rules of evidence, revisability, testability, and
coherence with historical and current ideas in science. While the NSES attempt a greater
level of articulation than the Benchmarks, there is still no guidance for science educators,
Science Education
curriculum developers, or other stakeholders regarding what gives a scientific explanation

its explanatory power and little guidance for helping science educators craft learning
environments that encourage construction and evaluation of scientific explanations.
The concept of scientific explanation plays a central role in Taking Science to School:
Learning and Teaching Science in Grades K-8 (NRC, 2007) and in its companion volume
for classroom teachers, Ready, Set, Science! (NRC, 2008). According to TSTS, students
should
• Know, use, and interpret scientific explanations of the natural world.

• Generate and evaluate scientific evidence and explanations.
• Understand the nature and development of scientific knowledge.
• Participate productively in scientific practices and discourses (p. 334)
TSTS draws upon developments in science studies to conceptualize science as a field that
is “fundamentally about establishing lines of evidence and using the evidence to develop and
refine explanations”; however, TSTS does not provide much guidance about the composition
of scientific explanations. Scientific explanations in TSTS are typically conceptualized as
“statistical models of natural phenomena . . . rooted in probabilistic reasoning” rather than
cause–effect relationships (p. 18), but further guidance for science educators is not provided.
The authors are critics of overly simplistic causal and mechanistic models for scientific
explanation and are advocates for statistical and probabilistic models, but these models
are not explicitly described. The volume critiques practices in science education that have
conceptualized learning as the acquisition of facts or knowledge of specific cause–effect
relationships and suggests shifting emphasis away from accumulation of facts toward a
focus on scientific explanations for natural phenomena.
Scientific explanation plays an integral role in on-going science education reform ef-
forts in Europe particularly in the United Kingdom. In Beyond 2000: Science Education
for the Future (1998), the authors propose a turn toward science education that fosters
understanding of the major “explanatory stories” of the natural sciences. This turn is in
reaction to the observation that much of science in schools was being presented as a static
body of knowledge irrelevant for students and for their futures as informed citizens.2 Like
the reform documents from the United States, Beyond 2000 offers broad recommendations
that science curricula should shift away from emphasis on describing and memorizing facts
and toward explanations, discussions, problem solving, and writing to communicate ideas.
Beyond 2000 details some of the “explanatory stories” suggested as key components of
science curricula and provides some guidance about the attributes of scientific explana-
tions. For example, explanations often stretch beyond available data employing creativity
when theorizing about how events happen because explanations are models of what people
think is happening at a level that is not directly observable. Beyond 2000 takes a position
that the power of an explanation is increased when that explanation can predict outcomes,
but concedes that sometimes there is not enough data to provide reliable predictions so in
certain cases science is unable to offer more than a correlation.
Stemming from the recommendations in Beyond 2000, a new experimental course was
developed for teachers and students as part of the science education reform efforts in the
United Kingdom. Breaking the Mould? Teaching Science for Public Understanding (2002)
offers an analysis of this reform effort and highlights some of the ways in which a turn
toward scientific explanation as envisioned by broad reform documents can be challenging
2
For a thoughtful critique of some of the framing of the Beyond 2000 report, please see Donnelly (2005).
Science Education
for teachers. Despite changes within the structure and content of the course, teachers’
classroom practice continued to rely heavily on presenting and explicating science ideas
for students rather than engaging students in first-hand examination of data, discussion of
ideas, or construction of scientific arguments and explanations. The authors attribute this
shortcoming to three interrelated challenges: (1) underprepared and unsupported teachers
who do not have pedagogical techniques for engaging in this kind of science instruction,
(2) a lack of pedagogical tools within the new curriculum, and (3) a lack of supports to
enable teachers to break away from the deeply ingrained practices of traditional science
instruction. Breaking the Mould recommends that future efforts at science education reform
should provide teachers with (1) support developing their own background knowledge about
major explanatory science ideas across science subdisciplines and about the ways in which
arguments and explanations are constructed in science, (2) support for providing scaffolding
to foster student-to-student discourse about science ideas, and (3) support for scaffolding
students written arguments and scientific explanations.
Recently, some of the concepts from Beyond 2000 and Breaking the Mould resurfaced in
recommendations for science education throughout Europe in Science Education in Europe:
Critical Reflections (2008). Notable among the recommendations in this document is a push
for science education across the European Union “to educate students both about the major
explanations for the material world that science offers and about the way science works”
(p. 15). Echoing recommendations of Breaking the Mould, this document stresses the need
for a shift in science teaching practices from “deductive to inquiry-based methods” and
help for teachers as they learn to engage in practices supporting student discussions and
writing evidence-based explanations and arguments (p. 23).
Reform documents call for more emphasis on scientific explanation in science class-
rooms. They push educators to move away from science education practices focused
on describing, measuring, and reporting about observable events or practices focused
on transmitting countless discrete facts to students. While these documents may help
in setting broad visions for science curricula and instruction, they do not provide the
level of specificity necessary to help teachers enact particular practices in their class-
rooms. Our recent efforts with teachers seek to fill this gap by developing conceptual
and pedagogical tools offering heuristic value for teachers to carry out specific instruc-
tional practices pressing for the co-construction of scientific explanations in science
classrooms (Thompson et al., 2009b, 2009c, 2010; Windschitl et al., 2008a,; 2008b,
2011).
CONSTRUCTING SCIENTIFIC EXPLANATIONS WITH TEACHERS

AND STUDENTS
When we imagine the practices that a science teacher uses to foster scientific explanation
with students, we see a distinction between practices focused on helping students make
sense of the explanations that they consume from authoritative sources like texts and teachers
and dialogic practices focused on having students produce—not just reproduce—scientific
explanations based on their own ideas and understanding of evidence. We have observed
that constructing and evaluating scientific explanations is a challenge both for science
students and for science teachers (Windschitl et al., 2008a, 2008b). Simply exhorting
teachers to engage in a practice—no matter how valuable—does not make the practice
happen. Teachers need scaffolding and support for learning how to engage in ambitious
science teaching practices, but as a field we are just beginning to understand those supports
for teacher learning. Along with our participating teachers, we have been wrestling with
Science Education
these issues for over 4 years, and in many cases the teachers are only now beginning to gain
insight into how the complex practice of scientific explanation can apply to the curriculum
they use every day (Thompson et al., 2009b, 2009c, 2010; Windschitl et al., 2008a, 2008b,
2011). For these teachers, as well as for us, pedagogical decision making about scientific
explanations in the classroom is not a straightforward task.
Five years ago, we began working with a cohort of novice teachers in graduate course-
work focused on model-based inquiry geared toward constructing and refining scientific
explanations with secondary students in Grades 6–12 (see Windschitl & Thompson, 2006,
and Windschitl et al., 2008b, for details about the design of these courses). We thought
we had guided our participants through a rich learning experience in the science teaching
methods course, and we were really looking forward to seeing how our novice teachers
engaged in ambitious science teaching in the classroom. After 20 weeks of coursework,
our participants began a 10-week teaching practicum. Initially teachers struggled to enact
the classroom practices necessary to engage their students in constructing evidence-based
scientific explanations. We immediately realized that the Methods course had not fully sup-
ported their learning and started to create new supports. By working to support our novice
teachers, we have deepened our own understanding of the complex practice of scientific
explanation. Here we share an overview of our intellectual growth as well as some of the con-
ceptual and pedagogical tools that are the tangible products emerging from 5 years of work
with teachers. For more detailed accounts of our recent research, please see Thompson
& Windschitl (2006), Thompson et al. (2009b, 2010), and Windschitl et al. (2008a,
2008b).
Conceptual and Pedagogical Tools for Scientific

Explanations in Classrooms
In response to our beginning teachers’ struggles, we realized that we had to provide
additional support in the form of conceptual and pedagogical tools—tangible tools and
practices that would foster particular socioprofessional practices that, until this moment,
we had vigorously espoused but had not facilitated. Our first move was the most critical: We
explicitly defined for ourselves and for our novice teachers what attributes characterized
“good” scientific explanations. We created a rubric—the Explanation Tool—delineating a
simplified continuum of scientific explanations that (1) employ major scientific theories, (2)
seek underlying theoretical causes for observable events in nature, and (3) when appropriate,
utilize mathematical models to describe patterns in data (see Table 3). This Explanation Tool
served as a powerful heuristic helping to organize our teachers’ thoughts about scientific
explanations in their classrooms.
We chose to use a mixture of models for explanation for two important reasons. First,
it is clear from both the philosophy of science literature and from studies of scientists at
work that multiple forms of scientific explanation are employed in the actual practices
of scientists. Second, to capitalize on teachers’ and students’ existing everyday epistemic
and discursive practices, we found it necessary to create a variety of connections between
everyday ways of reasoning and communicating and the specific practices associated with
scientific explanation. This is a pragmatic approach to explanation similar to the view of
Van Fraassen, but we also emphasize two models of explanation frequently: (1) causal
explanations and (2) unificationist explanations. Our rationale for this emphasis is three-
fold. First, we are trying to foreground some of the commonalities of scientific explanation
inherent in philosophy of science regardless of differences between subdisciplines. Second,
engaging in the practice of constructing these explanatory models fosters sense making
Science Education
662
TABLE 3
Explanation Tool Used to Help Teachers Envision and Evaluate the Depth of Students’ Explanations
Science Education
for learners. Finally, the process of exploring a trajectory from describing “what” hap-
pened to explaining “how” and “why” events happen helps teachers focus their science
curriculum on central, core ideas in science (Thompson et al., 2009c; Windschitl et al.,
2011).
The Explanation Tool offered teachers a decidedly oversimplified framework for thinking
about scientific explanations, but it also served an important heuristic purpose scaffolding
teacher learning in three ways. First, the Explanation Tool was used during conversations
with the cohort of teachers to propel teachers to examine the science content of their
lessons. Some teachers began to critique lessons in their standard instructional materials
because these lessons seemed to be missing key theoretical concepts or tended to focus ex-
clusively on measuring and describing easily observed variables (Thompson et al., 2009b,
2009c). Second, teachers used the Explanation Tool to refine the questions that they posed
to students during class and in written assignments. Some teachers used the Explanation
Tool to help sort through the myriad questions in their instructional materials and selected
questions that emphasized explanatory reasoning instead of questions asking students to
recall facts or define words (Thompson et al., 2010; Windschitl et al., 2011). Third, teach-
ers used the Explanation Tool to examine the possible lines of reasoning emerging from
their middle school and high school students. Some teachers used the Explanation Tool
to map out their learning goals in advance, and then to systematically analyze their stu-
dents’ responses throughout a unit of instruction tracing lines of thought and modifying
subsequent instruction (Thompson et al., 2010; Windschitl et al., 2011). The Explanation
Tool is not a tool to be used in isolation. Socioprofessional practices such as discussions
with other science educators, the systematic examination of students’ unfolding science
ideas, and analyses of the instructional supports given to students coupled with a power-
ful heuristic like the Explanation Tool pushes teacher practice forward (Windschitl et al.,
2011).
Teachers’ growth continued during the first year of professional teaching as their engage-
ment with the Explanation Tool during collaborative inquiry group meetings continued. A
second tool—a learning progression for teachers—was developed to help participants re-
flect on their own learning and their classroom teaching practices (see the Appendix). This
learning progression, similar in format to a rubric, outlines continua of development across
four areas, which were focal ideas during the teachers’ graduate teacher education program.
This tool enabled reflective conversations between teachers, colleagues, and instructional
coaches where teachers could reenvision their teaching practice and take steps to enact
increasingly ambitious pedagogy (Thompson et al., 2009b, 2010). Interestingly, teachers
who had struggled to engage in ambitious teaching practices at the beginning of the school
year noted that the focus on scientific explanations helped them find a “back door” into
model-based inquiry and helped them to locate the “big ideas” in their curricula that could
serve as the focus of their instruction (Thompson et al., 2009b, 2010). Previously, both
Model-Based Inquiry and the process of transforming mundane curriculum topics into big
science ideas had been too daunting for many teachers to attempt consistently during stu-
dent teaching and during the early months of professional teaching. A focus on scientific
explanations, supported by conceptual and pedagogical tools embedded in the practice of
collaborative inquiry with colleagues, seems to have served as an “on ramp” for many of
our teachers, allowing them access to ambitious pedagogy, which otherwise would have
been out of reach for many novices.
Our findings indicate that the tools served at least four valuable functions for teachers.
The tools (1) embody a valued practice (e.g., scientific explanation), (2) are applicable
across grade levels and subject matter subdomains, (3) represent practice in accessible
language, and (4) provide descriptions of levels of performance from which teachers and
Science Education
students can identify “where they are” and what constitutes the next levels of performance
(Windschitl et al., 2011). The heuristic functions of conceptual tools for teaching offer
science teachers and teacher educators an opportunity to reify ideas about what counts
in science classrooms. The use of tools for teaching has enabled us to have important
conversations about scientific practice and knowledge construction that have pushed our
teachers’ thinking as well as our own thinking about scientific explanations. We now are
focusing on a system of conceptual and pedagogical tools embedded in collegial practices
with like-minded peers to support teachers’ development of ambitious science teaching
practices. Our current and future work with preservice and early career science teachers
explores the possibility that with well-conceptualized systems of tools and with the constant
support of a group of colleagues, novice science teachers can accomplish ambitious forms
of science teaching that are otherwise generally unattainable by novices and by many expert
teachers as well.
SYNTHESIS
Science education reform literature makes numerous calls for teachers to create learning
environments emphasizing scientific explanations and argument construction. But, without
clearer, more articulated conceptualizations of explanation and argumentation, science
educators do not have the guidance necessary to make these visions of reform a reality
because to do so means assembling a repertoire of teaching practices that is currently well
beyond the norms of practice in classrooms. We hope to spark conversation across the
science education community regarding three questions: (1) What constitutes a “good”
scientific explanation in science classrooms?, (2) What makes an explanation explanatory
rather than descriptive?, and (3) How might we evaluate the merits of alternate explanations
offered by students in classrooms?
The science education community has made some attempts at addressing the first two
questions. Allusions to particular models of scientific explanation can often be found within
the reform literature, but to readers who are not well versed in philosophy of science these
details may go unnoticed. The field of science education can look to philosophy of science
for helpful models of scientific explanation. In particular, we suggest focusing on five
philosophical models of scientific explanation that are commonly part of students’ and
teachers’ everyday ways of engaging in explanation of natural phenomena, which may
mean opening a door to employing multiple models of explanation (see Table 1). While
there is much disunity in the sciences about concepts like scientific explanation, there are
some points of consensus that can help answer some of our questions. By explicitly engaging
with the range of possible models of explanation and making a principled choice for the
kind of scientific explanation desired in particular cases for school science, researchers
may be in a better position to clarify salient features of the epistemic practices involved for
teachers and for students.
We have found that the single most powerful conceptual tool for advancing science
teachers’ practice is to provide a way for teachers to distinguish between descriptive and
explanatory endeavors in science. Moving from an emphasis on “what” happens in the
natural world toward and emphasis on both “how” and “why” events happen opens new
trajectories for teachers working to develop ambitious repertoires of instructional practice in
science classes. Teachers who help students develop explanatory models based on evidence
and on major theoretical ideas in science can create learning environments where students
engage in productive disciplinary discourse throughout an entire school year developing
and refining “what counts” as a good theory, model, explanation, or argument together as
a community.
Science Education
Recently, we have uncovered a major challenge in helping teachers realize this vision.
We know that for teachers to create learning environments where students engage in model-
based reasoning it is important for teachers to work out for themselves the standards for
explanatory power, adequacy of explanations, and the strength of argument construction.
Using conceptual tools, like the Explanation Tool (see Table 3), teachers can engage in
important intellectual work together by building and refining scientific explanations for
phenomena central to the school science subjects that they teach. However, it is difficult
for teachers who have completed this intellectual work to then follow students’ ideas as
they emerge during classroom discourse. Instead, teachers work to bring students “on
board” with the teachers’ ideas. This is a perennial tension between authoritative and
dialogic approaches and continues to be worth serious examination by scholars in science
education.
It is the third question—How do we evaluate the merits of alternate explanations offered
by students in classrooms?—that seems to be most absent from discussion about scientific
explanations in school science. When our research group discusses learning environments
that press for scientific explanation, we are envisioning classrooms where teachers and
students work together to co-construct scientific explanations in the form of explanatory
models combining students’ ideas, science ideas, and the available evidence. But much of
the literature about scientific explanation in schools tends to focus on enhancing teachers’
ability to clearly communicate well-established scientific explanations to students who will
subsequently absorb and internalize them. This vision of science learning stands in stark
contrast to the learning environments envisioned in science education reforms, which tend
to prioritize shared knowledge construction in the science classroom.
Teachers who create learning environments where students are engaging in the process
of constructing and refining scientific explanations for themselves will be confronted by
situations where students have constructed a plausible, but ultimately incorrect, scientific
explanation. These “alternative explanations” present a tough pedagogical dilemma. In
science classrooms, it can be difficult, if not impossible at times, to provide students
and teachers with sufficient access to theory and evidence to allow for reasoning through
alternative explanations to ultimately arrive at an understanding consistent with current
scientific thinking. Moreover, there are often very good reasons for students to hold tightly
to plausible, even if ultimately incorrect, explanations for natural events. Science teachers
we have observed, including the first author, frequently reflect that there are many days
when students’ alternative explanations push teachers into a conceptual and pedagogical
“corner” where it seems that the only way out is to tell the student that her explanation is
incorrect and supply her with the correct explanation. This is not a desirable pedagogical
move, but we have yet to develop adequate pedagogical tools to help teachers work with
students’ plausible-but-incorrect scientific explanations.
One of the challenges faced by science teachers who attempt to build scientific ex-
planations with students is that both teachers and students often construct explanations
from partial understanding of science ideas, limited access to empirical data, and a wide
range of tacitly held ideas about the natural world. We would like the science education
community to engage in conversation about the fine line between engaging in sense mak-
ing about established scientific explanations and building explanatory models in science
classrooms—explanations that may ultimately include threads of partial understanding but
also threads of alternative conceptions. Much work remains to be done to help teachers and
students engage productively in the practice of constructing, testing, refining, and justifying
evidence-based scientific explanations, and we look forward to further conversations in the
field.
Science Education
APPENDIX: TEACHER’S PERFORMANCE PROGRESSION FOR MODEL-BASED INQUIRY 666
Science Education
We are grateful for the guidance from our research group at the University of Washington and the
local science teachers with whom we collaborate. This article benefitted from critical feedback from
three anonymous reviewers, the editor of this journal, Andrea Woody, Jessica Thompson, and Nathan
Murphy.
REFERENCES
Achinstein, P. (1983). The nature of explanation. New York: Oxford University Press.
American Association for the Advancement of Science. (1993). Benchmarks for science literacy. New York:
Oxford University Press.
Banilower, E. R., Smith, P. S., Weiss, I. R., & Pasley, J. D. (2006). The status of K-12 science teaching in the United
States: Results from a national observation study. In D. W. Sunal & E. L. Wright (Eds.), The impact of state
and national standards on K-12 science teaching (pp. 83 – 122). Greenwich, CT: Information Age Publishing.
Bell, P. (May 2010). Reflections on the mangle of science learning research. Paper presented at the annual meeting
of the American Educational Research Association, Denver, CO.
Berland, L. K., & Reiser, B. J. (2009). Making sense of argumentation and explanation. Science Education, 93,
26 – 55.
Berthold, K., Eysink, T. H. S., & Renkl, A. (2009). Assisting self-explanation prompts are more effective than
open prompts when learning with multiple representations. Instructional Science, 37(4). 345 – 363.
Braaten, M. (April 2010). Lenses on learning: The complex roles of tools and colleagues in a science teacher
video club. Paper presented at the annual meeting of the American Educational Research Association, Denver,
CO.
Bricker, L., & Bell, P. (2008). Conceptualizations of argumentation from science studies and the learning sciences
and their implications for the practices of science education. Science Education, 92, 473 – 498.
Cartwright, N. (1997). The truth doesn’t explain much. In D. Rothbart (Ed.), Science, reason, and reality: Issues
in the philosophy of science (pp. 161 – 166). Fort Worth, TX: Harcourt Brace.
Chinn, C., & Malhotra, B. (2002). Epistemologically authentic inquiry in schools: A theoretical framework for
evaluating inquiry tasks. Science Education, 86(2), 175 – 218.
Coleman, E. B. (1998). Using explanatory knowledge during collaborative problem solving in science. Journal of
the Learning Sciences, 7(3&4), 387 – 427.
Donnelly, J. (2005). Reforming science in the school curriculum: A critical analysis. Oxford Review of Education,
31(2), 293 – 309.
Driver, R., Asoko, H., Leach, J., Mortimer, E., & Scott, P. (1994). Constructing scientific knowledge in the
classroom. Educational Researcher, 23(7), 5 – 12.
Driver, R., Leach, J., Millar, R., & Scott, P. (1996). Young people’s images of science. Buckingham, England:
Open University Press.
Duschl, R. (1990). Restructuring science education: The importance of theories and their development. New York:
Teachers College Press.
Duschl, R. (2000). Making the nature of science explicit, In R. Millar, J. Leach, & J. Osborne (Eds.), Improving
science education: The contribution of research (pp. 187 – 206). Philadelphia: Open University Press.
Duschl, R., & Grandy, R. (Eds.). (2008). Teaching scientific inquiry: Recommendations for research and imple-
mentation. Rotterdam, The Netherlands: SensePublishers.
Erduran, S., & Jimenez-Aleixandre, M. P. (Eds.). (2008). Argumentation in science education: Perspectives from
classroom-based research. Berlin: Springer.
Friedman, M. (1974). Explanation and scientific understanding. Journal of Philosophy, 71, 5 – 19.
Gamoran, A., Anderson, C., Quiroz, P., Secada, W., Williams, T., & Ashmann, S. (2003). Transforming teaching
in math and science: How school districts can support change. New York: Teachers College Press.
Grotzer, T. A. (2003). Learning to understand the forms of causality implicit in scientifically accepted explanations.
Studies in Science Education, 39, 75 – 102.
Grotzer, T. A., & Basca, B. B. (2003). How does grasping the underlying causal structures of ecosystems impact
students’ understanding? Journal of Biological Education, 38(1), 16 – 29.
Hammer, D., Russ, R., Mikeska, J., & Scherr, R. (2008). Identifying inquiry and conceptualizing abilities, In
R. Duschl & R. Grandy (Eds.), Teaching scientific inquiry: Recommendations for research and implementation
(pp. 138 – 156). Rotterdam, The Netherlands: SensePublishers.
Harrison, A. G., & Treagust, D. F. (2000). Learning about atoms, molecules, and chemical bonds: A case study
of multiple-model use in grade 11 chemistry. Science Education, 84, 352 – 381.
Hempel, C. G. (1965). Aspects of scientific explanation. In C. G. Hempel (Ed.), Aspects of scientific explanation
and other essays in the philosophy of science. New York: Free Press.
Science Education
Hempel, C. G., & Oppenheim, P. (1948). Studies in the logic of explanation. Philosophy of Science, 15(2),
135 – 175.
Herrenkohl, L. R. (2006). Intellectual role taking: Supporting discussion in heterogeneous elementary science
classes. Theory into Practice, 45(1), 47 – 54.
Kitcher, P. (1989). Explanatory unification and the causal structure of the world. In P. Kitcher & W. C. Salmon
(Eds.), Minnesota studies in the philosophy of science: Vol. XIII. Scientific explanation (pp. 410 – 499).
Minneapolis: University of Minnesota Press.
Kitcher, P. (1997). Explanatory unification. In D. Rothbart (Ed.), Science, reason, and reality: Issues in the
philosophy of science (pp. 167 – 186). Fort Worth, TX: Harcourt Brace.
Kitcher, P., & Salmon, W. C. (Eds.). (1989). Minnesota studies in the philosophy of science: Vol. XIII. Scientific
explanation. Minneapolis: University of Minnesota Press.
Knorr-Cetina, K. (1999). Epistemic cultures: How sciences make knowledge. Cambridge, MA: Harvard University
Press.
Latour, B., & Woolgar, S. (1986). Laboratory life: The construction of scientific facts (2nd ed.). Princeton, NJ:
Princeton University Press.
Lemke, J. (1990). Talking science: Language, learning, and values. Norwood, NJ: Ablex.
McNeill, K. L. (2009). Teachers’ use of curriculum to support students in writing scientific arguments to explain
phenomena. Science Education, 93, 233 – 268.
McNeill, K. L., & Krajcik, J. (2008). Scientific explanations: Characterizing and evaluating the effects of teachers’
instructional practices on student learning. Journal of Research in Science Teaching, 45(1), 53 – 78.
McNeill, K. L., Lizotte, D. J., Krajcik, J., & Marx, R. W. (2006). Supporting students’ construction of scientific
explanations by fading scaffolds in instructional materials. Journal of the Learning Sciences, 15(2), 153 –
191.
Millar, R., & Osborne, J. F. (Eds.). (1998). Beyond 2000. Science education for the future. London: King’s College
London.
Moje, E. B., Peek-Brown, D., Sutherland, L. M., Marx, R. W., Blumenfeld, P., & Krajcik, J. (2004). Explaining
explanations: Developing scientific literacy in middle-school project-based science reforms. In D. Strickland
& D. E. Alvermann (Eds.), Bridging the gap: Improving literacy learning for preadolescent and adolescent
learning in grades 4 – 12. New York: Teachers College Press.
Mortimer, E., & Scott, P. (2003). Meaning making in secondary science classrooms. Philadelphia: Open University
Press.
Nagel, E. (1979). The structure of science: Problems in the logic of scientific explanation. Indianapolis, IN:
Hackett Publishing.
National Research Council. (1996). National Science Education Standards. Washington, DC: The National
Academies Press.
National Research Council. (2007). Taking science to school: Learning and teaching science in grades K-8.
Committee on Science Learning, Kindergarten through Eighth Grade. R. A. Duschl, H. A. Schweingruber, &
A. W. Shouse (Eds.), Board on Science Education, Center for Science Education, Division of Behavioral and
Social Sciences and Education. Washington, DC: The National Academies Press.
National Research Council (2008). Ready, set, science! Putting research to work in K-8 science classrooms. Sarah
Michaels, Andrew W. Shouse, & Heidi A. Schweingruber (Eds.), Board on Science Education, Center for
Science Education, Division of Behavioral and Social Sciences and Education. Washington, DC: The National
Academies Press.
Norris, S. P., Guilbert, S. M., Smith, M. L., Hakimelahi, S., & Phillips, L. M. (2005). A theoretical framework for
narrative explanation in science. Science Education, 89, 535 – 563.
Ogborn, J., Kress, G., Martins, I., & McGillicuddy, K. (1996). Explaining science in the classroom. Buckingham,
England: Open University Press.
Osborne, J. (May 2010). The importance of the history, philosophy, and sociology of science for the practice
of school science. Paper presented at the annual meeting of the American Educational Research Association,
Denver, CO.
Osborne, J., & Dillon, J. (2008). Science education in Europe: Critical reflections. London: King’s College.
Osborne, J., Duschl, R., & Fairbrother, R. (2002). Breaking the mould? Teaching science for public understanding.
London: King’s College.
Palincsar, A. S., Anderson, C., & David, Y. M. (1993). Pursuing scientific literacy in the middle grades through
collaborative problem solving. The Elementary School Journal, 93(5), 643 – 658.
Perkins, D. N., & Grotzer, T. A. (2005). Dimensions of causal understanding: The role of complex causal models
in students’ understanding of science. Studies in Science Education, 41, 117 – 166.
Roth, K., & Garnier, H. (2007). What science teaching looks like: An international perspective. Educational
Leadership, 64(4), 16 – 23.
Science Education
Russ, R. S., Scherr, R. E., Hammer, D., & Mikeska, J. (2008). Recognizing mechanistic reasoning in student
scientific inquiry: A framework for discourse analysis developed from philosophy of science. Science Education,
92(3), 499 – 525.
Salmon, W. C. (1978). Why ask “Why?” An inquiry concerning scientific explanation. Proceedings and Addresses
of the American Philosophical Association, 51(6), 683 – 705.
Salmon, W. C. (1989). Four decades of scientific explanation. Minneapolis: University of Minnesota Press.
Sandoval, W. A. (2003). Conceptual and epistemic aspects of students’ scientific explanations. Journal of the
Learning Sciences, 12(1), 5 – 51.
Sandoval, W. A., & Millwood, K. A. (2005). The quality of students’ use of evidence in written scientific
explanations. Cognition and Instruction, 23(1), 23 – 55.
Sandoval, W. A., & Reiser, B. J. (2004). Explanation-driven inquiry: Integrating conceptual and epistemic scaffolds
for scientific inquiry. Science Education, 88, 345 – 372.
Songer, N., Kelcey, B., & Gotwals, A. (2009). How and when does complex reasoningoccur? Empirically driven
development of a learning progression focused on complex reasoning about biodiversity. Journal of Research
in Science Teaching, 46(6), 610 – 631.
Thompson, J., Braaten, M., & Windschitl, M. (June 2009b). Learning progressions as vision tools for advancing
teachers’ pedagogical performance. Paper presented at Learning Progressions in Science (LeaPS) conference,
Iowa City, IA.
Thompson, J., Braaten, M., & Windschitl, M., with Sjoberg, B., Jones, M., & Martinez, K. (2009c). Examining
student work: Evidence-based learning for students and teachers. The Science Teacher, 76(8), 48 – 52.
Thompson, J., Windschitl, M., & Braaten, M. (April 2008). How pedagogical reasoning and practice develops
across “learning-to-teach” contexts. Paper presented at annual meeting of the American Educational Research
Association, New York.
Thompson, J., Windschitl, M., & Braaten, M. (April 2009a). Explaining the development of ambitious practices
across “learning-to-teach” contexts. Paper presented at the annual conference of the National Association of
Research in Science Teaching. Anaheim, CA.
Thompson, J., Windschitl, M., & Braaten, M. (March 2010). Critical and contextual discourses: Explaining
the development of ambitious practices across “learning-to-teach” contexts. Paper presented at the annual
conference of the National Association for Research in Science Teaching, Philadelphia, PA.
Traweek, S. (1988). Beamtimes and lifetimes: The world of high energy physicists. Cambridge, MA: Harvard
University Press.
Treagust, D. F., Chittleborough, G., & Mamiala, T. L. (2003). The role of submicroscopic and symbolic represen-
tations in chemical explanations. International Journal of Science Education. 25(11), 1353 – 1368.
Treagust, D. F., & Harrison, A. G. (2000). In search of explanatory frameworks: An analysis of Richard Feynman’s
lecture “Atoms in motion.” International Journal of Science Education, 22(11), 1157 – 1170.
Van Fraassen, B. C. (1980). The scientific image. Oxford, England: Clarendon.
Windschitl, M. (2008). What is inquiry? A framework for thinking about authentic scientific practice in the
classroom. In J. Luft, R. Bell, & J. Gess-Newsome (Eds.), Science as inquiry in the secondary setting
(pp. 1 – 20). Arlington, VA: NSTA press.
Windschitl, M., & Thompson, J. (2006). Transcending simple school science investigations: Can pre-service
instruction foster teachers’ understandings of model-based inquiry? American Educational Research Journal,
43(4), 783 – 835.
Windschitl, M., Thompson, J., & Braaten, M. (2008a). How novice science teachers appropriate epistemic
discourses around model-based inquiry for use in classrooms. Cognition and Instruction, 26(3), 310 – 378.
Windschitl, M., Thompson, J., & Braaten, M. (2008b). Beyond the scientific method: Model-based inquiry as a
new paradigm of preference for school science investigations. Science Education, 92, 941 – 967.
Windschitl, M., Thompson, J., & Braaten, M. (2011). Ambitious pedagogy by novice teachers: Who benefits from
tool-supported collaborative inquiry into practice and why? Teachers College Record, 113(7).
Windschitl, M., Thompson, J., Braaten, M., Stroupe, D., Chew, C., & Wright, E. (April 2010). The beginner’s
repertoire: A core set of instructional practices for teacher preparation. Paper presented at the annual meeting
of the American Educational Research Association, Denver, CO.
Wittwer, J., & Renkl, A. (2008). Why instructional explanations often do not work: A framework for understanding
the effectiveness of instructional explanations. Educational Psychologist, 43(1), 49 – 64.
Science Education

Braaten 2011

Uploaded by

Copyright:

Available Formats

Braaten 2011

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Braaten 2011

Uploaded by

Copyright:

Available Formats

SCIENCE STUDIES AND SCIENCE EDUCATION

Sibel Erduran and John L. Rudolph, Section Coeditors

Working Toward a Stronger

Received 11 May 2010; revised 11 November 2010; accepted 29 November 2010

Correspondence to: Melissa Braaten; e-mail: mbraaten@uw.edu

1. How is scientific explanation defined within philosophy of science?

FIVE MODELS OF SCIENTIFIC EXPLANATION FROM PHILOSOPHY

The “Covering Law” Model of Scientific Explanation

Covering Law Explanations in Science Classrooms. Critiques from Salmon and

about explanations by using logical argument construction based on persistent patterns

Statistical Model of Scientific Explanation

when a carcinogen causes cancer. Examining data to arrive at a statistical or probabilistic

Statistical/Probabilistic Explanations in Science Classrooms. In school science con-

Causal Model of Scientific Explanation

Causal Explanations in Science Classrooms. Many of the phenomena examined in

Pragmatics of Explanation in Science Classrooms. Understanding Van Fraassen’s ar-

Explanatory Unification View of Scientific Explanation

Explanatory Unification in Science Classrooms. The explanatory unification view of

Summary of Models of Explanation

CONCEPTUALIZATIONS OF SCIENTIFIC EXPLANATION

Scientific Explanations in Science Education Research

Explanation as Explication. In the science education literature, it is common to see

• Explain what biological evolution that they are working on scientific

Explanation as Causation. Some scholars in science education take a perspective that

students to create a particular kind of scientific explanation—a causal explanation using

Explanation as Justification. Scientific explanation and argumentation are key practices

Scientific Explanations in Science Education Reform Documents

curriculum developers, or other stakeholders regarding what gives a scientific explanation

• Know, use, and interpret scientific explanations of the natural world.

CONSTRUCTING SCIENTIFIC EXPLANATIONS WITH TEACHERS

Conceptual and Pedagogical Tools for Scientific

You might also like