13

Science education

What are the benefits for scientists and others from learning the general prin-
ciples of scientific method? And which scholars have best clarified and docu-
mented those benefits?
The first chapter mentioned various positive or negative assessments of
these benefits by scientists and philosophers. But it noted that the most careful
assessments have been by science educators, while leaving documentation of
their findings to this chapter. Six benefits were listed: better comprehension,
greater adaptability, greater interest, more realism, better researchers, and
better teachers.
Science educators typically place various aspects of scientific method within
several broader contexts, including inquiry, practice, and the nature of science
(NOS). “For science educators the phrase ‘nature of science,’ is used to describe
the intersection of issues addressed by the philosophy, history, sociology, and
psychology of science as they apply to and potentially impact science teaching
and learning. As such, the nature of science is a fundamental domain for guiding
science educators in accurately portraying science to students” (McComas,
Clough, and Almazroa, in McComas 1998:5). Educators and scientists alike
perceive the importance of the NOS in the science curriculum (McComas 1998;
Khine 2011).
This chapter first reviews the typical NOS concepts that have been developed
primarily for K–12 (kindergarten through high school) science education. It
then reviews numerous and extensive empirical investigations by science edu-
cators that document the previous six benefits. Finally, it lists 10 academic NOS
concepts better suited to undergraduate and graduate students and science
professionals.

Typical NOS concepts

The principal approach to teaching the NOS, particularly in the context of

K–12 education, has been to develop a consensus view that reflects mainstream

234
Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Typical NOS concepts 235

Typical NOS Concepts for K–12 Education

Scientific knowledge while durable has a tentative character.

Scientific knowledge relies heavily but not entirely, on observation, experimental
evidence, rational arguments, and skepticism.
There is no one way to do science (therefore, there is no universal step-by-step
scientific method).
Science is an attempt to explain natural phenomena.
Laws and theories serve different roles in science; therefore students should note that
theories do not become laws even with additional evidence.
People from all cultures contribute to science.
New knowledge must be reported clearly and openly.
Scientists require accurate record keeping, peer review, and replicability.

Observations are theory-laden.

Scientists are creative.
The history of science reveals both an evolutionary and revolutionary character.
Science is part of social and cultural traditions.
Science and technology impact each other.

Scientific ideas are affected by their social and historical milieu.

Figure 13.1 Fourteen typical concepts for the nature of science (NOS) for K–12 education.
This list represents a consensus among science education guidelines from eight nations.

science. The National Research Council of the National Academy of Sciences

has expressed its view that “Although there is no universal agreement about
teaching the nature of science, there is a strong consensus about characteristics
of the scientific enterprise that should be understood by the educated citizen”
(NRC 2012:78). Figure 13.1 shows 14 NOS concepts derived from a com-
parison of 8 international science education standards, including the United
States, Canada, United Kingdom, New Zealand, and Australia (McComas
et al., in McComas 1998:6–7). Likewise, the science content standards for all 50
US states incorporate these NOS elements, although with varying comprehen-
siveness (William F. McComas, Carole K. Lee, and Sophia J. Sweeney, personal
communication).
Partly because the NOS is a mandated topic and partly because it is inherently
of great interest to scientists and educators, assessment instruments have been
developed. Although more than 40 instruments have been developed, the Test
on Understanding Science (TOUS) and the Views of Nature of Science (VNOS,

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 236 Science education

in several versions for different grade and developmental levels) instruments

have been especially widely used (Lederman 2007). Widely adopted instru-
ments facilitate comparisons among the numerous empirical studies that have
used them. Nevertheless, there are some legitimate questions about consensus
concepts for the NOS.
An important cultural question is, “Is science universal?” (Cobern and
Loving 2001). This is a fair question, given that the NOS standards and assess-
ment instruments mentioned herein represent a rather homogeneous group
of English-speaking nations. There are perplexing and sensitive issues regard-
ing Western and indigenous perspectives on science. Some indigenous beliefs
have no potential whatsoever to be of interest to mainstream science, such
as the example that Cobern and Loving gave of some South Pacific islanders
who attribute tides to great sea turtles leaving from or returning to their nests,
rather than to the gravitational tug of the moon (and sun). On another topic
such as drug discovery, both mainstream science and indigenous knowledge
have contributed substantially. On yet another topic, such as identifying local
crops with potential for future development into major crops by conventional
and molecular plant breeding, the contributions from indigenous cultures are
essential. Cobern and Loving advised that indigenous knowledge be “valued for
its own merits” without confusing or diluting the methodology and knowledge
of mainstream science.
A practical question regards whether NOS concepts are general, equally
applicable across all sciences, or else are context specific, variously applicable
in astronomy or biology or sociology or whatever. For instance, Irzik and
Nola (2010) argued for a family resemblance approach to the NOS rather
than a consensus approach: “Science is so rich and so dynamic and scientific
disciplines are so varied that there seems to be no set of features that is common
to all of them and shared only by them.” They argued that this alternative is
more comprehensive because it “captures the elements of NOS described by
the consensus view” while it also “contains novel elements” and thereby “does
justice to the differences among scientific disciplines.” But recall from Chapter
1 that the American Association for the Advancement of Science expressed a
nuanced position on commonality and diversity among the various scientific
disciplines (AAAS 1989:25–26, 29). Accordingly, a family resemblance among
the various sciences may be regarded as a component of the consensus view
rather than as a competitor to it.
To address this issue about the relative importance of general and context-
specific aspects of NOS concepts, Urhahne, Kremer, and Mayer (2011) con-
ducted a large empirical study, surveying 221 secondary-school students in
Germany. The students’ NOS conceptions were assessed for seven concepts
named: Source, Certainty, Development, Justification, Simplicity, Purpose,
and Creativity. The 10 specific contexts were: Smoking, Mobile phone, Cli-
mate change, Aggression, Intelligence, Dinosaurs, Continental drift, Big Bang,

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Typical NOS concepts 237

Out of Africa, and Evolution. The students’ general NOS conceptions were
assessed with a 40-item questionnaire and their context-specific NOS concep-
tions by a 10-item questionnaire for each of the 10 specific contexts. Table 7
in Urhahne et al. (2011) presented the 70 correlation coefficients between gen-
eral and context-specific NOS conceptions for the 7 concepts in the 10 specific
contexts.
For this concepts-by-contexts dataset, the relative importance of general and
context-specific aspects of the NOS can be quantified by a standard statistical
tool, the analysis of variance. It identifies three sources of variance: concept
main or general effects, context main effects, and concept-by-context inter-
action effects. These account for 1%, 26%, and 73% of the total variance,
respectively. Hence, interaction effects (73%) seem to be about three times as
important as main effects (27%). However, to be more precise, only half of this
interaction contains systematic patterns of interest, and the remainder is mostly
noise. Hence, general and context-specific aspects of NOS were about equally
important in this particular study.
Such results make sense. Although researchers in the physical, biological, and
social sciences make use of all of the general principles of scientific method, the
emphasis on various principles does vary somewhat among disciplines.
As evidence that a treatment of scientific method can serve diverse scientists
and nonscientists alike, despite some variations in emphases, I would men-
tion the tremendous breadth of disciplines that have publications citing this
book’s predecessor, Scientific Method in Practice (Gauch 2002). Those publica-
tions are in many sciences: anthropology, astronomy, bioinformatics, chemistry,
climatology, ecology, geophysics, hydrology, morphology, natural resources,
oceanography, physics, proteomics, psychology, seismology, sociology, taxon-
omy, toxicology, and zoology. They are also in technology: agriculture, com-
puter science, engineering, medicine, pharmacology, public health, and space
exploration. And they are in business, economics, education, law, management,
music, and philosophy.
In review, there is considerable consensus on basic NOS concepts, and there
are widely adopted assessment instruments. Furthermore, this NOS consen-
sus suits multicultural settings, while respecting indigenous knowledge for its
own merits, but without confusing or diluting mainstream science. There is
evidence that the NOS has context-specific as well as general aspects, so differ-
ent scientific disciplines such as geology and physics have somewhat different
NOS emphases.
However, there are legitimate concerns about characterizations of scientific
method in the typical NOS concepts developed for K–12 education. “The issue
of methodology seems to be dismissed altogether by saying that there is no
single method for doing science. . . . While it is certainly true that there is no
single scientific method in the sense of a mechanical procedure that determines
knowledge production step by step, there are general methodologies . . . that

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 238 Science education

guide scientific practice in general ways. Moreover, without the idea of a sci-
entific method or methodological rule, it is difficult to see how science can be
self-corrective and provide reliable knowledge” (Irzik and Nola 2010). Indeed,
the third item in Figure 13.1 is “There is no one way to do science (there-
fore, there is no universal step-by-step scientific method),” without any further
elaboration on scientific method. Accordingly, a later section in this chapter
proposes more advanced concepts for the NOS in general and scientific method
in particular that are suitable for undergraduate and graduate students and sci-
ence professionals. But, first, the following six sections document benefits that
result from studying and mastering the general principles of scientific method.

Better comprehension

The NRC has expressed the essential goal of science education in a mere four
words: “scientific knowledge with understanding” (NRC 1996:21). Hence, sci-
entific literacy requires comprehension of both scientific content and scientific
method. This literacy also relates to technological competence.
An explicit goal of the National Science Education Standards is to establish high levels
of scientific literacy in the United States. An essential aspect of scientific literacy is
greater knowledge and understanding of science subject matter, that is, the knowledge
specifically associated with the physical, life, and earth sciences. Scientific literacy also
includes understanding the nature of science, the scientific enterprise, and the role of
science in society and personal life. . . .
The goal of science is to understand the natural world, and the goal of technology
is to make modifications in the world to meet human needs. . . . Technology and sci-
ence are closely related. A single problem often has both scientific and technological
aspects. (NRC 1996:21–24)

Students must master scientific method and the NOS in order to make
scientific knowledge their own: “As Plato insisted so long ago, education is
not just the having of correct beliefs, it is the having of adequate reasons for
these beliefs” (Matthews 1998). To describe the opposite situation for a student
learning only science’s facts, he quoted memorable words from John Locke:
“Such borrowed wealth, like fairy money, though it be gold in the hand from
which he received it, will be but leaves and dust when it comes to use.”
One item among the “modest goals” for NOS instruction advocated by
Matthews (1998) is logic: “Given that being able to reason clearly is an obvious
component of being scientific, then one of the low-level nature of science
objectives might be to teach some elementary formal and informal logic.” Basic
instruction in deductive and inductive logic can reduce common fallacies and
misconceptions and can enhance reasoning, particularly the kind of reasoning
needed to evaluate competing hypotheses.

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Better comprehension 239

Of course, in general, both reasoning ability and prior knowledge of a sub-

ject help students comprehend more advanced material, and yet surprisingly
the former is often more important. For instance, what factors influence stu-
dents’ ability to comprehend college biology? Johnson and Lawson (1998)
conducted an empirical study of 366 students enrolled in a nonmajors biol-
ogy course at a community college. They pretested students to determine their
reasoning ability and prior knowledge. Then some classes with about half of
the students received “expository instruction” focused on facts and concepts,
whereas the other classes received “inquiry instruction” giving more attention
to science’s logic, method, and process. The results were that “Reasoning ability
but not prior knowledge or number of previous biology courses accounted for
a significant amount of variance in final examination score in both instruc-
tional methods.” Furthermore, the inquiry classes showed better improvement
in both reasoning skills and biology comprehension.

Given that reasoning ability appears to be a significant predictor of achievement in

introductory level college biology, more so than the number of previous biology courses
completed or a pretest measure of domain-specific knowledge, high school biology
instructors would be well advised to be more concerned with the development of their
students’ reasoning abilities than with making certain that they cover a wide range
of specific biology concepts. The same advice would also seem to be appropriate for
college level instructors who may be concerned about their students’ performance in
more advanced college courses. . . . The inquiry students in the present study not only
showed greater improvement in reasoning ability during the semester than the expository
students, they also did better on the measures of biology achievement. In other words,
nothing of importance seems to be lost by switching to inquiry instruction, and much
seems to be gained. (Johnson and Lawson 1998)

A similar win-win situation for scientific content and scientific method was
encountered in the context of college physics. Numerous students were tested
on acceleration, velocity, vectors, and forces.

It is well known that students come to the study of science with many naive prior notions
about the physical world and that these notions are very difficult to change. . . . It is less
well appreciated that students also come with many naive conceptions about the goals
of science and about the kinds of thinking needed for science. These conceptions,
imported from everyday life or from prior schooling, are even more difficult to change
than students’ naive notions about the physical world. Furthermore, their effects are
all-pervasive, affecting greatly what students try to learn and how they go about learning
it. . . .
An introductory physics course needs thus also to discuss explicitly the goals of science
and the ways of thinking useful in science. These issues cannot merely be addressed by
a few occasional remarks. They need to be constantly kept in students’ focus, and be
used as a framework within which more specific scientific knowledge and methods are
embedded. (Reif 1995)

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 240 Science education

Yet another win-win combination for content and method was found by
Cartier and Stewart (2000) in the context of high school genetics. “In contrast to
what is popularly believed, focusing on epistemological issues in the classroom
does not have to mean teaching less subject matter in exchange. In fact, helping
students to appreciate the epistemological structure of knowledge in a discipline
can provide them with the means to more fully understand particular knowledge
claims and their interrelationships” (Cartier and Stewart 2000).
These empirical findings prompt some commentary on the National Science
Education Standards (NRC 1996). The Standards include prominent curricu-
lum requirements for the NOS or, in their terminology, for “science as inquiry.”
That is commendable, given the findings of science educators. However, a table
on changing emphases is organized under two headings: “Less emphasis on”
and “More emphasis on” (NRC 1996:113). The very first item is less emphasis
on “Knowing scientific facts and information” and more emphasis on “Under-
standing scientific concepts and developing abilities of inquiry.” This might be
called the trade-off model, or the win-lose model, in which inquiry wins and
information loses. Admittedly, given a fixed schedule, an hour more for the
NOS does mean an hour less for science content. But that does not necessarily
imply less content because NOS instruction can pave the way for more rapid
coverage of more deeply comprehended material. By analogy, first sharpening
a dull ax means less time spent chopping wood, but that does not necessarily
mean less firewood and, indeed, the more likely outcome is more firewood.
Hence, a better model, and a realistic expectation as well, would be to put both
“Knowing scientific facts and information” and “Understanding scientific con-
cepts and developing abilities of inquiry” under the column “More emphasis
on.”
Similarly, the more recent NRC (2012:41) position paper on science education
reiterates this perception of tension or trade-off between content and method:
“From its inception, one of the principal goals of science education has been to
cultivate students’ scientific habits of mind, develop their capability to engage in
scientific inquiry, and teach them how to reason in a scientific context. There has
always been a tension, however, between the emphasis that should be placed
on developing knowledge of the content of science and the emphasis placed
on scientific practices” or methods. But, again, extensive empirical evidence
supports a positive interaction or synergy between content and method, not a
negative interaction or tension.

Greater adaptability

Four terms in the educational literature have closely interrelated meanings:

adaptability, versatility, creativity, and knowledge transfer. All four involve

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Greater adaptability 241

making the most of what is known by effectively connecting previous expe-

riences to new applications.
Science and technology are experiencing rapid and pervasive changes, requir-
ing scientists to be increasingly adaptable and flexible. Young scientists can
expect to change jobs and assignments more frequently than their predecessors.
They can also expect much of what they learn to have a short half-life of only a
few years. This has profound implications for science education.
Accordingly, the National Science Foundation (NSF 1996) and National
Academy of Sciences (NAS 1995) position papers for reshaping undergrad-
uate and graduate science education, respectively, promote adaptability as a
prominent goal. Regrettably, “We know that students rarely realize the appli-
cability of knowledge from one context to another” (NSF 1996:3). The first
of three general recommendations from the NAS (1995:76) is that “graduate
programs should add emphasis on versatility; we need to make our students
more adaptable to changing conditions.” Versatility or adaptability has become
so essential that the NAS recommends that evaluation criteria for funding edu-
cation/training grants “include a proposer’s plan to improve the versatility of
students” (NAS 1995:80).
Lawson et al. (2000) found that hypothesis-testing skills helped even more
than subject-matter knowledge in solving a new “transfer” problem. They
had a large sample of 667 college undergraduates. One predictor variable was
hypothesis-testing skills, assessed by a 13-item written test with each answer
scored as correct or incorrect. The other predictor variable was subject-matter
knowledge, assessed by a five-item multiple-choice test concerning momen-
tum, air, helium, and balloons. The dependent variable was the score (correct
or incorrect) for a new or transfer problem about a video showing the contrast-
ing motions of a balloon filled with air and another with helium in a moving
vehicle that came to an abrupt stop. Multiple regression analysis determined
which predictors were significant. “The analysis revealed that hypothesis-testing
skills, but not declarative [subject-matter] knowledge, accounted for a signif-
icant amount of variance” in the dependent variable, success in solving the
balloon transfer problem. Also, across four levels of increasing hypothesis-
testing skills, the percentages of students giving correct answers on the transfer
problem rose consistently: 17%, 34%, 57%, and 65%.
Educators and psychologists have discovered ways to increase adaptability or
knowledge transfer (Lobato 2006). What is crucial for transfer is not merely the
number of examples shown but also the explicit and reflective attention given
to general principles. In transfer problems, students tend to notice primarily
the surface features held in common, whereas they must be encouraged to look
beyond surface features to shared general or causal principles. That is, to solve
new problems, students need to abstract, discounting the superficial details of
examples and noticing the general principles. A particularly helpful strategy is
comparing cases (Gentner, Loewenstein, and Thompson 2003).

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 242 Science education

The importance of knowledge transfer is especially acute in medicine because

such transfer is precisely what diagnosis amounts to: applying past learning to
the patient at hand. For example, Ark, Brooks, and Eva (2007) found that
explicit attention on reasoning strategies increased the ability of medical stu-
dents to identify key features in electrocardiograms and assign correct diagnoses.
Although an extensive literature from educators and psychologists provides
strategies for promoting knowledge transfer, there is evidence that much of this
insight has not yet informed the pedagogy of most medical professors (Laksov,
Lonka, and Josephson 2008).
Precisely because the general principles of scientific method are general,
whereas specialized research techniques are specialized, these general principles
are that portion of science that has the greatest potential for frequent and
pervasive applications. Hence, general principles have extraordinary relevance
for adaptability, versatility, creativity, and knowledge transfer. Nevertheless,
this tremendous potential is not actualized automatically but rather requires an
informed and disciplined implementation of effective pedagogy.

Research clearly shows . . . that transfer of NOS knowledge does not happen automat-
ically. Students do not learn relevant NOS aspects through historical examples alone
or by instruction that refers only casually to elements of the nature of science. Several
researchers have pointed out that complex NOS ideas should be accompanied by explicit
and reflective discussion of the underlying concepts and principles. (Urhahne et al. 2011)

NOS knowledge can become transfer knowledge, not inert knowledge, by

accompanying examples with explicit and reflective discussion, by comparing
rather than merely multiplying examples, and by emphasizing general prin-
ciples over superficial similarities. The most obvious aspect of adaptability or
transfer is application of familiar knowledge to new and similar cases. How-
ever, knowledge transfer also concerns preparation for later learning of more
advanced material, as well as reconsideration and enrichment of previously
learned situations in light of subsequent examples and insights.

Greater interest

As in many nations, stimulating greater student interest in science and technol-

ogy is seen as a national priority for the United States.

The understanding of, and interest in, science and engineering that its citizens bring
to bear in their personal and civic decision making is critical to good decisions about
the nation’s future. The percentage of students who are motivated by their school and
out-of-school experiences to pursue careers in these fields is currently too low for the

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Greater interest 243

nation’s needs. Moreover, an ever-larger number of jobs require skills in these areas,
along with those in language arts and mathematics. . . .
A rich science education has the potential to capture students’ sense of wonder about
the world and to spark their desire to continue learning about science throughout their
lives. Research suggests that personal interest, experience, and enthusiasm—critical to
children’s learning of science at school or other settings—may also be linked to later
educational and career choices. Thus, in order for students to develop a sustained
attraction to science and for them to appreciate the many ways in which it is pertinent
to their daily lives, classroom learning experiences in science need to connect with their
own interests and experiences. (NRC 2012:x, 28)

Accordingly, it is important to observe that science educators have found

that science in its historical and philosophical context has greater appeal for
most students than does science abstracted from its context. “Incorporating
the nature of science while teaching science content humanizes the sciences
and conveys a great adventure rather than memorizing trivial outcomes of the
process. The purpose is not to teach students philosophy of science as a pure
discipline but to help them be aware of the processes in the development of
scientific knowledge” (William F. McComas, in McComas 1998:13).
For example, Becker (2000) produced a sequence of “history-based” science
lessons particularly to reach “those students identified as traditionally alienated
from the world of science and technology.” She reported that “Students who
took the course found it satisfying, diverse, historical, philosophical, humani-
tarian, and social. They . . . found the historical approach to be interesting and
the text enjoyable to read.” The material engaged “learners of varying abili-
ties” from a “socio-economically, ethnically, and linguistically diverse student
population.”
Another example is the Harvard Project Physics course, which incorporates
some history and philosophy of science. More than 60 studies of the effectiveness
of that course have been published, all positive: “Measures such as retention [of
students] in science, participation by women, improvement on critical think-
ing tests and understanding of subject matter all showed improvement where
the Project Physics curriculum was adopted” (Matthews 1994:6). However,
although it increased interest in science, it was not effective for communicating
the NOS.
In attempting to stimulate student interest in science, it helps to know that
science educators have partitioned student interest into two main components
termed “individual” and “situational” interest (Seker and Welsch 2006). They
are quite different. Individual interest is “stable interest, enduring over time and
characterized by high levels of stored knowledge and stored value” on the part
of a given person. By contrast, situational interest “is generated primarily by
certain conditions and/or concrete objects . . . in the environment” and it “tends

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 244 Science education

to be short-lived, lasting only as long as the situation provides interest.” It can be

stimulated by puzzles, games, stories about the personal lives of scientists, role-
playing, drama writing, movies, and other forms of entertainment. Obviously,
long-term individual interest is more important than ephemeral situational
interest. However, if situational interest is stimulated for extended periods of
time, it may eventually promote individual interest.
Students with high individual interest and those with low individual interest
had considerably different reactions to different instructional approaches. Stu-
dents with high interest reacted positively to material on the history and nature
of science. But others with low interest “expected to memorize what the science
teacher told them and what the textbook emphasized” and disliked historical
review showing that ideas develop over time and even are sometimes corrected
or discarded. For instance, one student complained, “If someone is going to
come along in the next few years and come up with something new, then what’s
the point? I’m not learning this stuff if it will change.”
A particularly important part of the NOS for interesting a diversity of stu-
dents in science is its simplest, common-sense elements. Warren et al. (2001)
contrasted perspectives on the relationship between science and common sense
as being either “fundamentally discontinuous” or “fundamentally continu-
ous” and found the latter much more conducive to learning science: “Our
work . . . shows that children, regardless of their national language or dialect,
use their everyday language routinely and creatively to negotiate the com-
plex dilemmas of their lives and the larger world. Likewise, in the science
classroom children’s questions and their familiar ways of discussing them do
not lack complexity . . . or precision; rather, they constitute invaluable intel-
lectual resources which can support children as they think about and learn
to explain the world around them scientifically.” Culturally sensitive “every-
day sensemaking” is a powerful bridge to scientific interests and patterns of
thought.
Although the skills involved in understanding the scientific method and
the NOS promote understanding science, at a deeper level, one’s interest in
cultivating those skills must be promoted by a desire to seek the truth. An
important study by Ben-Chaim, Ron, and Zoller (2000) explored numerous
skills and dispositions related to scientific thinking. They reported that “truth-
seeking is courageous intellectual honesty, a major attribute of CT [critical
thinking], which, in turn, is an important component of HOCS [higher-order
cognitive skills].”

More realism

Realism about science requires understanding the accuracy, confidence, and

scope of any given scientific result. But this topic inherently and integrally

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 More realism 245

involves scientific method because methods precede and enable results. The
level of certainty accorded to any given scientific result ranges from highly
tentative to confident fact, and the scope ranges from particular situations to
great generality.

Science is replete with ideas that once seemed promising but have not withstood the test
of time, such as the concept of the “ether” or the vis vitalis (the “vital force” of life). Thus
any new idea is initially tentative, but over time, as it survives repeated testing, it can
acquire the status of a fact—a piece of knowledge that is unquestioned and uncontested,
such as the existence of atoms. . . .
Scientific knowledge is a particular kind of knowledge with its own sources, justifications,
ways of dealing with uncertainties, and agreed-on levels of certainty. When students
understand how scientific knowledge is developed over systematic observations across
multiple investigations, how it is justified and critiqued on the basis of evidence, and
how it is validated by the larger scientific community, the students then recognize that
science entails the search for core explanatory constructs and the connections between
them. They come to appreciate that alternative interpretations of scientific evidence can
occur, that such interpretations must be carefully scrutinized, and that the plausibility
of the supporting evidence must be considered. . . .
Decisions must also be made about what measurements should be taken, the level
of accuracy required, and the kinds of instrumentation best suited to making such
measurements. As in other forms of inquiry, the key issue is one of precision—the goal
is to measure the variable as accurately as possible and reduce sources of error. The
investigator must therefore decide what constitutes a sufficient level of precision and
what techniques can be used to reduce both random and systematic error. . . .
Under everyday circumstances, . . . Newton’s second law accurately predicts changes in
the motion of a single macroscopic object of a given mass due to the total force on it.
But the second law is not applicable without modification at speeds close to the speed of
light. Nor does it apply to objects at the molecular, atomic, and subatomic scale or to an
object whose mass is changing at the same time as its speed. (NRC 2012: 79, 251, 59–60,
114)

Realism about science requires a discerning grasp of science’s powers and

limits. As already documented in Chapter 6, awareness of science’s powers and
limits has been widely recognized as a key component of scientific literacy
(AAAS 1989:26, 1990:20–21; NRC 1996:21). For science and technology under-
graduates, the NRC (1999:34) posed six specific questions that they should be
able to answer: “How are the approaches that scientists employ to view and
understand the universe similar to, and different from, the approaches taken
by scholars in other disciplines outside of the natural sciences? What kinds
of questions can be answered by the scientific and engineering methods, and
what kinds of questions lie outside of these realms of knowledge? How does
one distinguish between science and pseudoscience? Why are scientists often
unable to provide definitive answers to questions they investigate? What are

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 246 Science education

risk and probability, and what roles do they play when one is trying to provide
scientific answers to questions? What is the difference between correlation and
causation?”
What science educators have found is that an understanding of science’s
method powerfully promotes the educational goal of developing realistic ideas
about science. This makes sense given that powers and limits are largely con-
sequences of methods: “Understanding how science operates is imperative for
evaluating the strengths and limitations of science” (William F. McComas, in
McComas 1998:12). Also, “The ability to distinguish good science from paro-
dies and pseudoscience depends on a grasp of the nature of science” (Matthews
2000:326).
Empirical studies by science educators have shown that good understanding
of scientific method promotes realistic beliefs not only for ordinary scientific
inquiries but also for scientific matters having substantial worldview import.
For example, Lawson and Weser (1990) conducted a large empirical study of 954
college students in a nonmajors biology course. Scientific reasoning was assessed
by students’ answers to 16 questions involving conservation of mass; volume
displacement; control of variables; and reasoning about proportions, proba-
bilities, combinations, and correlations. Students who performed poorly were
referred to as “intuitive” thinkers, moderately well as “transitional” thinkers,
and quite well as “reflective” thinkers. Scientific beliefs were examined with
seven questions regarding special creation, orthogenesis, the soul, constitu-
tive nonreductionism, vitalism, teleology, and nonemergentism. Students were
pretested for both scientific reasoning and scientific beliefs, and then post-tested
after the biology course by readministering the same questionnaires. These data
were subjected to PCA to characterize the main patterns, as well as statistical
tests to detect specific differences between intuitive and reflective students.
Their main finding was: “As predicted, the results showed that the less skilled
reasoners were more likely to initially hold the nonscientific beliefs and were
less likely to change those beliefs during instruction. It was also discovered
that less skilled reasoners were less likely to be strongly committed to the
scientific beliefs.” Their suggested explanation was that reflective students hold
beliefs more strongly than intuitive students “because they have acquired the
reasoning skills that enable them to consider the alternative beliefs and the
evidence in a hypothetico-deductive manner to arrive at firm conclusions.”
There were numerous positive interactions between reasoning level and strength
of confidence in scientific beliefs. Their two main conclusions were: “(1) teachers
should be aware that students may hold beliefs that are inconsistent, even
contradictory to scientific beliefs and that these beliefs may be difficult to
alter; (2) some students lack the reasoning skills that appear to be necessary
to comprehend the arguments and evidence in favor of scientific beliefs; thus
instruction should focus on ways of improving student reasoning skills as well
as teaching scientific conceptions.”

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Better researchers 247

Better researchers

Research competence for undergraduate and graduate science majors is a major

educational objective. Again, my thesis is that the winning combination for sci-
entists is strength in both the research technicalities of a given specialty and
the general principles of scientific method. Hence, it is virtually axiomatic that
a deep understanding of scientific method develops better researchers. Fur-
thermore, what has already been documented in this chapter regarding better
comprehension, adaptability, interest, and realism makes for better researchers.
Ryder and Leach (1999) conducted an empirical investigation of research
competence among science undergraduates in the UK as a function of their
understanding of the general principles of scientific method. They found that
such understanding “can have a major impact on students’ activities during
investigative [research] projects,” specifically on “students’ decisions about
whether to repeat an experiment, whether to question an earlier interpretation
of their data as a result of new evidence, where to look for other scientific work
related to their project and what counts as a conclusion to their investigation.”
On the other hand, inadequate or naı̈ve conceptions of scientific method “con-
strain student learning” and made it impossible for students to achieve some of
their research objectives. More exactly, they found that mere diligence sufficed
for students to function as competent technicians, collecting accurate data by
means of a prescribed protocol; but real understanding of scientific method was
needed for students to function as creative scientists, understanding how data
and theory interact to support knowledge claims.
Interestingly, they investigated not only the students but also the students’
research supervisors. For one student with a particularly naı̈ve and limiting
view of scientific method, they found no evidence that the student’s supervisor
recognized the problem or implemented any strategy to remedy the deficiency.
Professors and supervisors must realize that teaching more and more about
research technicalities is not the required remedy when the real problem is lack
of understanding of some basic principle of scientific inquiry. Clearly, problems
of the latter sort are quite common.
Hunter, Laursen, and Seymour (2007) conducted an empirical study of sum-
mer undergraduate research at four liberal arts colleges with a sample of 138
students, 71 faculty, and 9 administrators. They found that typical undergrad-
uate research prompts quite limited growth in NOS understanding. “Many
faculty and student observations reported gains in applying their knowledge
and skills to research work, although fewer mentioned increases in higher
order thinking skills, particularly the development of a complex epistemolog-
ical understanding of science or the ability to define a research question and
develop experimental design.” The process of developing a real understanding
of the NOS by means of undergraduate research experience “is neither easy

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 248 Science education

nor guaranteed.” This result echoes Ryder and Leach’s conclusion that it is
easier to produce competent technicians than creative scientists. It also rein-
forces the unanimous finding in countless studies that progress in understand-
ing the NOS, even in the context of authentic research experience, will not occur
unless this topic receives intentional, explicit, reflective attention.
Sadler et al. (2010) provided an ambitious review of 53 empirical studies
of research apprenticeships for secondary and college students as well as K–12
teachers. Research apprenticeships promoted positive outcomes for interest and
career aspirations in science, knowledge of specific science content, technical
and statistical skills, confidence in one’s ability to do scientific research, and
dispositions for effective communication and collaboration. By contrast, on the
matter of principal concern here, again extensive empirical evidence showed
that research experience prompted rather little growth in NOS understanding.

Apprenticeship programs are often touted as effective vehicles for learning about the
NOS. The basic argument is that opportunities to work with scientists offer new per-
spectives on how science is done and the nature of scientific knowledge. This argument
is based on an implicit model of NOS learning. That is, learners will gain sophisti-
cated understandings of NOS by virtue of participating in authentic science activities
without instruction dedicated to the realization of this goal. Many science education
researchers have challenged the notion that implicit approaches for teaching about NOS
results in meaningful learning. These critics have argued that explicit approaches that
support learner development of sophisticated NOS ideas through targeted instruction
and learner reflection are more likely to result in NOS learning. It is possible to partner
apprenticeship experiences with explicit NOS instruction. (Sadler et al. 2010)

Of their 53 studies, 20 reported on NOS learning, with 19 relying on implicit

models and only 1 implementing explicit instruction. Schwartz, Lederman, and
Crawford (2004) interviewed 13 preservice teachers involved in a course that
combined a research setting with explicit NOS instruction. The VNOS instru-
ment was used to assess NOS understanding. In stark contrast with the miserable
performance of implicit models, this explicit approach resulted in 11 of its 13
participants gaining in NOS understanding. Explicit instruction, rather than
research apprenticeship alone, was the key influence. More recently, Bautista
and Schussler (2010) implemented explicit NOS instruction successfully within
an introductory college biology laboratory.
In review, extensive empirical studies by science educators support two prin-
cipal findings about the relationship between NOS understanding and research
competence. First, greater understanding of the NOS in general and scientific
method in particular does develop better researchers, especially by clarifying how
data and theory interact to support knowledge claims. Second, regarding the
reverse direction, research experience does not easily or automatically increase
NOS understanding but rather progress with the NOS requires explicit and
reflective instruction, whether or not that instruction occurs within the context

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Better teachers 249

of scientific research. Science educators should bear in mind these two findings
as they seek to give scientists their winning combination of strength in both the
research technicalities of a given specialty and the general principles of scientific
method.

Better teachers

Regarding science and technology faculty, the NSF (1996:ii) advises that “It
is important to assist them to learn [and teach] not only science facts, but,
just as important, the methods and processes of research.” Likewise, the NRC
(1996:21) requires that science teachers understand the NOS, including the
“modes of scientific inquiry, rules of evidence, ways of formulating questions,
and ways of proposing explanations,” and the NRC (1999:41, 2012:253–260,
319–323) reaffirms those requirements. In addition to items such as “physical
science” and “life science,” one of the eight categories of content standards for
science is the “history and nature of science” (NRC 2001:148).
Matthews (1994:199–213, 2000:321–351) documented that knowledge of
the history, philosophy, and method of science can strengthen teachers. Such
knowledge “can improve teacher education by assisting teachers to develop a
richer and more authentic understanding of science, . . . can assist teachers [to]
appreciate the learning difficulties of students, because it alerts them to the his-
toric difficulties of scientific development and conceptual change, . . . [and] can
contribute to the clearer appraisal of many contemporary educational debates
that engage science teachers and curriculum planners” (Matthews 1994:7). The
NRC concurs.
Teaching science . . . requires that teachers have a strong understanding of the scientific
ideas and practices they are expected to teach, including an appreciation of how scientists
collaborate to develop new theories, models, and explanations of natural phenomena.
Rarely are college-level science courses designed to offer would-be science teachers, even
those who major in science, the opportunity to develop these understandings. Courses
designed with this goal are needed. (NRC 2012:256)

But unfortunately, science teachers often lack a sophisticated and consistent

understanding of scientific method and inquiry (Abd-El-Khalick and Lederman
2000; Lederman 2007). Naturally, teacher weaknesses in the NOS have direct
implications for student weaknesses. “What is . . . clear from recent research is
that rarely do students receive instruction in the K–12 grades that contribute
to better understandings of NOS unless their teachers have had some kind of
professional development for teaching NOS” (Akerson et al. 2011).
But the good news is that professional development can work. “Sustained
professional development improves teachers’ abilities to teach science, and
recent research shows that professional development can improve teachers’

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 250 Science education

conceptualizations of nature of science (NOS) to those in line with recommen-

dations of national reforms as well as their abilities to explicitly teach NOS to
their own elementary students” (Cullen, Akerson, and Hanson 2010). Indeed,
as one might well expect, there is extensive empirical evidence that high-quality
teaching increases student achievement (NRC 2001:44–65).
Nevertheless, extensive empirical evidence also shows that teacher NOS
knowledge does not automatically translate into student NOS knowledge.
“Research on the translation of teachers’ conceptions into classroom prac-
tice, however, indicates that even though teachers’ conceptions of NOS can be
thought of as a necessary condition, these conceptions, nevertheless, should
not be considered sufficient” (Lederman 2007). Pedagogy and curriculum are
additional factors. To learn and then to teach the NOS, explicit and reflective
attention on this topic is needed.
Increasingly, many nations and states are mandating that science educators
take courses in the NOS (Matthews 2000:321–351). Also there is a trend toward
mandating that assessments of scientific literacy include questions about the
NOS (Norman G. Lederman et al., in McComas 1998:331–350). But such man-
dates must be implemented carefully: “I have become more convinced . . . that
improving evaluation of science instruction is probably the most important
part of the equation for improving science education. But, if evaluative pro-
cesses and materials are not built upon the best knowledge about the nature
of science and applied appropriately for the developmental level of students,
through school and university, they will fall short of what is needed” (Robinson
1998).
Finally, the best motivation for science teachers to learn and teach the NOS is
awareness of demonstrated benefits. As already mentioned, national standards
for science education, mandated assessments of scientific literacy, and other
factors are sure to increase the attention given to the NOS. But for science
educators, true motivation, as contrasted with perfunctory compliance, arises
from known benefits. Indeed, “reports indicating positive changes in students’
views and actions regarding the nature of science are needed to bolster teachers’
confidence that attention to these issues will reap the desired effects” (William
F. McComas et al., in McComas 1998:29). Accordingly, this chapter’s documen-
tation of six benefits from explicit and reflective study of the NOS and scientific
method has great relevance for motivating teacher interest in the NOS, and
thereby student interest also.

Academic NOS concepts

The importance of science literacy is widely appreciated, especially in the K–12

curriculum. But Moore (1998), then-president of the scientific research society

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Academic NOS concepts 251

Sigma Xi, asked the incisive question, “But what kind of scientific literacy is
important?” His reply was that “The basic concepts measured by survey ques-
tions may have little to do with the kinds of knowledge that are needed for
competitiveness in the technological world.” Accordingly, it would be regret-
table if “Basic literacy and numeracy take precedence over more sophisticated
scientific understanding in the priorities of education.” The very real value of
basic scientific literacy for “simple good citizenship in a technological world”
for all K–12 students should not be allowed to supersede or dwarf the equally
real value of advanced scientific literacy for college students and science profes-
sionals. Neither elementary nor advanced education should be neglected.
Accordingly, in contrast to the typical NOS concepts for K–12 education that
Figure 13.1 showed, Figure 13.2 shows academic NOS concepts suitable for
college students and science professionals. There is almost no overlap between
these two lists of concepts.
These ten academic NOS concepts have two main sources. One is the litera-
ture in science, statistics, and philosophy, which informs topics such as the PEL
model, deductive and inductive reasoning, and parsimony and accuracy. The
other is position papers on science education, which motivate topics such as
the relationship between the sciences and the humanities, science’s presuppo-
sitions, and science’s worldview import. These 10 items are representative, not
exhaustive. Additional topics in this book on scientific method include expli-
cation and defense of science’s rationality, science’s ethics and responsibilities,
and effective pedagogy for science education.
Between Figures 13.1 and 13.2, there is one intentional and direct conflict:
The first item in the typical NOS concepts is “Scientific knowledge while durable
has a tentative character,” whereas the corresponding last item in the academic
NOS concepts is “Much scientific knowledge is certain, some is probable, and
some is speculative; and scientists ordinarily make careful, justified, and real-
istic claims for their findings.” In support for this shift toward more confident
language, there has been a change from the excessive emphasis on tentative-
ness in the AAAS (1989, 1990) and NRC (1996) position papers to explicit
recognition of “fact” and “knowledge that is unquestioned and uncontested”
in the more recent NRC (2012:79) position paper, as already quoted at greater
length in this chapter’s section on more realism. Indeed, “a focus on practices
(in the plural) avoids the mistaken impression that . . . uncertainty is a universal
attribute of science. In reality, practicing scientists employ a broad spectrum of
methods, and although science involves many areas of uncertainty as knowl-
edge is developed, there are now many aspects of scientific knowledge that are
so well established as to be unquestioned foundations of the culture and its
technologies. It is only through engagement in the practices that students can
recognize how such knowledge comes about and why some parts of scientific
theory are more firmly established than others” (NRC 2012:44).

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 252 Science education

Academic NOS Concepts

Science is one of the liberal arts. There is a rich traffic of ideas between the sciences
and the humanities. Philosophy of science is especially relevant for scientists.
Scientific inquiry involves the general principles of scientific method pervading all
sciences and the research techniques of a given specialty. General principles need be
mastered only once, whereas specialized techniques change from project to project.
The winning combination for scientists is mastering both.
The PEL model asserts that when fully disclosed, any argument supporting a scientific
conclusion necessarily has premises of three kinds: presuppositions, evidence, and
logic.
Science’s presuppositions are installed by philosophical reflection on any exemplar of
rudimentary common sense, such as “Moving cars are hazardous to pedestrians.”
Science’s presuppositions and logic are rooted in common sense, not worldview dis-
tinctives, so they are worldview independent. But empirical and public evidence from
the sciences and the humanities can have worldview import, and personal experience
is also potentially relevant although not public evidence. Worldview conclusions based
on worldview-independent presuppositions and logic and on worldview-informative
empirical and public evidence are the prerogative of individual scientists; but they are
not the prerogative of scientific institutions because the scientific community lacks
consensus.
Deduction reasons from a given model to expected observations. Predicate logic
requires only several axioms, and arithmetic, probability, and other deductive systems
such as geometry and calculus can be added with several more axioms.
Induction reasons from actual observations to inferred model. The probability of evi-
dence given hypothesis, P(E | H ), is related to its reverse conditional probability,
P(H | E ), by Bayes’s theorem, which also involves the prior probability P(H ). Con-
fusing these reverse conditional probabilities is a fallacy, as is ignoring the influence
of prior probabilities.
Evidence is evaluated by multiple criteria, including accurate fit with the data, par-
simony, robustness to noise, predictive accuracy, generality, unification of diverse
phenomena, explanatory power, testability, and coherence with other knowledge.
For noisy data being fitted by a model family, which represents an extremely common
situation in science and technology, the optimal tradeoff is at the peak of Ockham’s
hill, with less complex models underfitting signal and more complex models overfitting
noise. A parsimonious model can be more accurate than its data, thereby increasing
repeatability and accelerating progress.
Much scientific knowledge is certain, some is probable, and some is speculative; and
scientists ordinarily make careful, justified, and realistic claims for their findings.

Figure 13.2 Ten basic concepts for the nature of science (NOS) at an academic level.
University undergraduate and graduate students and science professionals need to
understand the NOS at a higher level than the typical NOS concepts developed for K–12
education.

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Academic NOS concepts 253

Reviews of the NOS literature make it abundantly clear that contempo-

rary science education is dominated by the opinion that all scientific knowl-
edge is tentative (Lederman 1992, 2007; Abd-El-Khalick and Lederman 2000).
Although atypical of most other science educators, Ernst von Glasersfeld (in
Tobin 1993:26) followed his perceptions of science’s woes to their logical con-
clusion of radical skepticism: “To conclude that, because we have a perceptual
experience which we call ‘chair,’ there must be a chair in the ‘real’ world is
to commit the realist fallacy. We have no way of knowing what is or could be
beyond our experiential interface.”
What explains such sentiments that all science (and perhaps all common
sense too) is tentative and revisable? Recall from Chapter 4 that there has
been a historical shift in philosophy of science from the excessively confident
philosophy begun by logical empiricists in the 1920s to the excessively diffident
philosophy emerging in the 1960s. That shift prompted the insightful remark
by Callebaut (1993:xv) that “Philosophy of science as currently practiced is a
reaction against a reaction” – a pendulum swinging from one extreme to the
opposite extreme. Correspondingly, there has been a shift in science education,
decade by decade from 1960 to 2000, from confident to tentative views of
scientific knowledge (Abd-El-Khalick and Lederman 2000). Accordingly, with
apologies to Callebaut, one might judge that science education as currently
practiced is a reaction against a reaction – whereas it ought to be a response
to nature and thereby a service to students and a foundation for technology.
Fortunately, the recent NRC (2012) position paper on science education insists
that uncertainty is not a universal attribute of science and, more pointedly, that
science possesses unquestioned and uncontested knowledge of unobservables
such as atoms.
Persistence in teaching the disingenuous opinion that all science is tentative
and revisable seems likely to have two deleterious side effects. First, just as too
much exposure to violence or entertainment tends to produce insensitivity,
a pervasive and incessant tentativeness causes unresponsiveness. By contrast,
the focused and occasional detection of specific blunders in actual scientific
inquiry does prompt response and correction. Second, telling students that
an official precept of the NOS is that all science is tentative and revisable,
when many of these same students perceive from their science courses that
numerous scientific facts are certain and unrevisable, constitutes an invitation
to confusion and insincerity. It sends the message that the NOS is in the realm
of silly philosophical games rather than actual scientific practices.
A promising direction for further research in science education is empirical
study of the benefits for college students and science professionals from studying
more advanced academic NOS concepts. Interesting questions include: When
scientists with and without having received academic NOS instruction are com-
pared, what differences emerge regarding research, teaching, and administrative

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 254 Science education

abilities? When a researcher’s awareness of the general principles of scientific

method (such as Ockham’s hill) enables an important innovation within his
or her own scientific specialty or technological application, what factors affect
how quickly it catches on? What makes for effective pedagogy and stimulat-
ing classroom dynamics when teaching academic NOS to undergraduate and
graduate students?
By their frequent use in numerous empirical studies, the TOUS and VNOS
instruments for assessing understanding of typical NOS concepts have helped to
standardize expectations and to facilitate comparisons among studies. Accord-
ingly, development of a corresponding instrument for academic NOS concepts
could be quite helpful. Items such as those in Figure 13.2 could be considered
for inclusion.
Finally, this chapter on science education concludes with an expression of
science education at its very best. By “liberal,” Matthews simply means encom-
passing the humanities.
At its most general level the liberal tradition in education embraces Aristotle’s delin-
eation of truth, goodness, and beauty as the ideals that people ought to cultivate in their
appropriate spheres of endeavor. That is, in intellectual matters truth should be sought,
in moral matters goodness, and in artistic and creative matters beauty. Education is to
contribute to these ends: it is to assist the development of a person’s knowledge, moral
outlook and behavior, and aesthetic sensibilities and capacities. For a liberal, education
is more than the preparation for work. . . . The liberal tradition seeks to overcome intel-
lectual fragmentation. Contributors to the liberal tradition believe that science taught
from such a perspective, and informed by the history and philosophy of the subject,
can engender understanding of nature, the appreciation of beauty in both nature and
science, and the awareness of ethical issues unveiled by scientific knowledge and created
by scientific practice. . . . The liberal tradition maintains that science education should
not just be an education or training in science, although of course it must be this, but also
an education about science. Students educated in science should have an appreciation of
scientific methods, their diversity and their limitations. They should have a feeling for
methodological issues, such as how scientific theories are evaluated and how competing
theories are appraised, and a sense of the interrelated role of experiment, mathematics
and religious and philosophical commitment in the development of science. . . . The
liberal approach requires a great deal from teachers; this needs to be recognized and
provided for by those who educate and employ teachers. (Matthews 1994:1–3, 6)

Summary

Science educators typically place various aspects of scientific method within

several broader contexts, including inquiry, practice, and the nature of sci-
ence (NOS). Comparisons among standards for K–12 science education, both
national and international, have produced a consensus view of the NOS. The

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 Summary 255

TOUS and VNOS instruments for assessing student understanding of the NOS
use that consensus, and their wide adoption has facilitated comparisons among
studies.
From hundreds of empirical studies, science educators have found six ben-
efits from explicit and reflective study of the NOS. (1) Understanding of sci-
entific method promotes learning of scientific content, so rather than being
competitive goals, there is a win-win situation for content and method. (2)
Understanding the general principles of scientific method promotes greater
adaptability, transferring knowledge to solve new problems. (3) A humanities-
rich version of science that includes historical and philosophical elements from
the NOS stimulates greater student interest in science. (4) Understanding scien-
tific method and the NOS promotes more realism about the accuracy and scope
of specific findings as well as the powers and limits of science more generally. (5)
Mastering scientific method is necessary for developing better researchers who
are not merely competent technicians but rather are actually creative scientists.
(6) Grasping the NOS results in better teachers of both the NOS and science
content.
College students and science professionals need to master academic NOS
concepts rather than the typical NOS concepts intended for K–12 students.
Ten basic items merit inclusion at the academic level. (1) Science is among
the liberal arts and there is a rich traffic of ideas between the sciences and the
humanities. (2) The winning combination for scientists is mastering both the
general principles of scientific method pervading all sciences and the research
techniques of a given specialty. (3) The PEL model asserts that when fully
disclosed, any argument supporting a scientific conclusion necessarily has
premises of three kinds: presuppositions, evidence, and logic. (4) Science’s
presuppositions are installed by philosophical reflection on any exemplar of
rudimentary common sense, such as “Moving cars are hazardous to pedestri-
ans.” (5) Although science’s presuppositions and logic are rooted in common
sense and thereby are worldview independent, empirical and public evidence
from the sciences and the humanities can have worldview import, and per-
sonal experience is also potentially relevant although not public evidence. (6)
Deduction reasons from a given model to expected observations. (7) Induction
reasons from actual observations to inferred model. (8) Evidence is evaluated
by multiple criteria, including accurate fit with the data, parsimony, robust-
ness to noise, predictive accuracy, generality, unification of diverse phenom-
ena, explanatory power, testability, and coherence with other knowledge. (9)
A parsimonious model, at the peak of Ockham’s hill, can be more accurate
than its data, thereby increasing repeatability and accelerating progress. (10)
Much scientific knowledge is certain, some is probable, and some is speculative;
and scientists ordinarily make careful, justified, and realistic claims for their
findings.

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015
 256 Science education

Study questions

(1) Regarding the first benefit of NOS instruction, better comprehension, what
do you think of the win-win situation depicted for NOS and science? Is
the empirical evidence convincing? Do you find this win-win situation
surprising or expected?
(2) What are some empirical findings by science educators showing that NOS
understanding promotes adaptability, interest, and realism?
(3) What are some empirical findings by science educators showing that NOS
understanding makes for better researchers and better teachers?
(4) What are three or four academic NOS concepts relevant for enhancing
perspective on science? Explain each briefly.
(5) What are three or four academic NOS concepts relevant for increasing
scientific productivity? Explain each briefly.

Downloaded from https://www.cambridge.org/core. University of Exeter, on 29 Dec 2017 at 16:19:35, subject to the Cambridge Core terms
of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139095082.015