This article was downloaded by: [Linköping University Library]

On: 04 October 2014, At: 02:56

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Science

Education
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/tsed20

RESEARCH REPORT
a a
Huann-Shyang Lin & Houn-Lin Chiu
a
Department of Chemistry , Ching-Yang Chou, Graduate Institute of
Science Education, National Kaohsiung Normal University , Taiwan
E-mail:
Published online: 22 Feb 2007.

To cite this article: Huann-Shyang Lin & Houn-Lin Chiu (2004) RESEARCH REPORT, International
Journal of Science Education, 26:1, 101-112, DOI: 10.1080/0950069032000070289

To link to this article: http://dx.doi.org/10.1080/0950069032000070289

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 INT. J. SCI. EDUC., 23 JANUARY 2004, VOL. 26, NO. 1, 101–112

RESEARCH REPORT

Student understanding of the nature of science and their

problem-solving strategies

Huann-Shyang Lin and Houn-Lin Chiu, Department of Chemistry, and

Ching-Yang Chou, Graduate Institute of Science Education, National
Downloaded by [Linköping University Library] at 02:56 04 October 2014

Kaohsiung Normal University, Taiwan; e-mail: t1666@nknucc.nknu.edu.tw

The purpose of this study was to examine the relationship between student understanding of the nature of
science and their problem-solving strategies. Six hundred and twenty Year 8 students in Taiwan twice completed
two conceptual problem-solving tests and a questionnaire on the nature of science. Four of these students were
selected for follow-up interviews. The result of stepwise multiple regression indicated that the subscale on the
nature of the scientific method consistently appeared as the best predictor for student problem-solving ability,
explaining about 22% of the variance. It was also found that problem-solving strategies were more conceptually
based for students that were high scorers on the nature of science survey.

Introduction
Promoting students’ problem-solving ability has long been considered one of the
most important goals of science education. Unfortunately, research studies have
found that many students perform algorithmic or mathematical manipulation by
rote memorization of formulas without having a basic understanding of specific
concepts (Gabel and Sherwood 1980, Nakhleh 1993). Furthermore, Nakhleh and
Mitchell (1993) found that college students’ conceptual problem-solving ability
lagged far behind their algorithmic problem-solving ability. They indicated that
some students retrieved complex chemical formulas, although the question only
required simple conceptual understanding. The factors affecting student problem-
solving ability and performance are many and varied. However, the overemphasis
on quantitative calculations in most science textbooks may be the cause of students’
low level of conceptual understanding. This argument can be exemplified by the
findings of De Berg (1989). His study analyzed 14 chemistry textbooks on the topic
of gas laws, and found that more than 93% of the examples and exercise problems
were quantitative while only 6.3% of them were qualitative. Because most teachers
rely very heavily on textbooks, it is reasonable to suspect that this emphasis has a
great impact on teachers’ instruction and assessment methods. No wonder most
students prefer the algorithmic problems that require mathematical calculations
even though they admit that conceptual problems are unique and helpful (Nakhleh
and Mitchell 1993). These findings have attracted the concern of science educators
and researchers. The literature review suggests that science teachers should pay
more attention to student conceptual problem-solving ability. Students should at
least be able to explore basic qualitative understanding before they are engaged in

International Journal of Science Education ISSN 0950–0963 print/ISSN 1464–5289 online © 2004 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080/0950069032000070289
 102 H.-S. LIN ET AL.

quantitative calculation.
The reasons why students find problem-solving difficult have been identified by
Frazer (1982), Osborne and Wittrock (1983), Selvaratnam (1983), (Gabel et al.
1984) and Lee (1985). The major reasons include students’ failure to construct
meaning from the problem statements, not being able to link the meaning of the
problem to their knowledge structure, or simply lacking the appropriate knowledge
structure for that specific content area. These studies all indicated the importance
of being able to construct and relate one’s personal knowledge structure to problem-
solving.
As important as problem-solving ability, understanding the nature of science
has been regarded as one of the basic requirements for scientific literacy. In Science
for all Americans (American Association for the Advancement of Science 1989), the
Downloaded by [Linköping University Library] at 02:56 04 October 2014

meaning and importance of the nature of science are extensively described. Duschl
(1990) argued that teachers’ view of scientific theory is likely to be related to their
method for selecting content and instruction strategy. Indeed, this positive
relationship has been further confirmed by Brickhouse (1990), Gallagher (1991),
and Lederman and Zeidler (1987). Although more consistent evidence must be
compiled in the future, science educators have reached a consensus that
understanding the nature of science is an important goal of science education.
Unfortunately, many research studies have found that students and teachers do not
have enough understanding of the nature of science (Aikenhead and Ryan 1991,
Carey and Stauss 1970, Mackay 1971, Rubba et al. 1981). Continuing efforts have
been underway since the 1960s to find the factors that affect and ways to improve
student levels of understanding in this regard (for example, Conant 1957, Klopfer
and Cooley 1963, Carey and Stauss 1970, Lin 1998, Solomon et al. 1992). It
should be noticed that, among these studies, using the history of science has been
regarded as a fruitful strategy.
The relationship between student science learning and their understanding of
the nature of science has caught the attention of science educators. It was found that
student views of science or epistemological beliefs are significantly related to their
knowledge integration and learning orientation methods (Songer and Linn 1991,
Tsai 1998). Students holding post-positivist-oriented views about science tended to
recall more information, show more flexibility, and demonstrate better met-
acognitive ability than students with empiricist-aligned epistemological beliefs. In
addition, previous studies have found that using the history of science was
successful in promoting student conceptual understanding (Lin 1998) and
understanding about the nature of science (Klopfer and Cooley 1963, Lin and
Chen 2002, Solomon et al. 1992). The entire picture seems to be that conceptual
understanding, the nature of science, and history of science together could be a
fruitful field for study. If we have a better knowledge of how student understanding
of the nature of science is related to their learning, this knowledge would be able to
serve as a guide or reference for those who are trying to integrate the history of
science into teaching. For instance, if we find that student knowledge of scientific
theory is important to their learning, curriculum designers can focus more on how
a scientific theory is discussed, debated and finally accepted by the scientific
community. The first purpose of this study was therefore to investigate what
subscales of the nature of science are the best predictors for student conceptual
problem-solving ability, and to compare the problem-solving strategies between
students holding post-positivist-oriented views and students with empiricist-aligned
 UNDERSTANDING THE NATURE OF SCIENCE AND PROBLEM-SOLVING 103

views. Second, historical materials were developed and will be tested in a pilot
program. Third, the effectiveness of these historical materials on promoting student
conceptual problem-solving ability will be investigated. This article reports on the
first part of the study.

Methodology
Instruments
Two conceptual problem-solving tests and one questionnaire on the nature of
science (NOS) were used in this study. The first test covered pressure, density,
buoyancy, and heat and temperature. The concepts included in the second test
were: atoms, molecules, atomic weights, molecular weights, and chemical reac-
Downloaded by [Linköping University Library] at 02:56 04 October 2014

tions. Both tests were pilot tested and have been used in previous studies by Lin
et al. (1996) and Lin (1998). The seven items in the first test produced a
Cronbach alpha of 0.74. The five items in the second test were 0.70 on the
reliability scale. All of the items in these two tests required students to apply
appropriate concepts to explain or predict a phenomenon. The content validity
was assured by asking two faculty chemists, two science educators, and two high
school science teachers to rate the degree of representation in the content
covered, the readability, and the clarity of the items. One sample item from each
of the two tests is shown in figures 1 and 2.
The NOS questionnaire consisted of 53 statement items on a five-point Likert
scale, with 1 standing for strongly agree and 5 standing for strongly disagree. These

Figure 1. The test item of density.

Figure 2. The test item of molecular weight.

 104 H.-S. LIN ET AL.

items were derived mainly from the Nature of Scientific Knowledge Scale (Rubba
and Anderson 1978). Two sample items are shown as follows.
1. When two scientists investigate a same natural phenomena, they are going
to present a similar report (empiricist-aligned item).
2. Today’s scientific laws, theories, and concepts might be changed in the face
of new evidence (post-positivist-oriented item).
Further evidence of the NOS questionnaire’s validity was obtained by interviewing
a subgroup of 20 students who took the questionnaire. It was concluded that their
answers in the interview questions were consistent with the responses on the
questionnaire. A similar approach to assessing the validity of the Likert-scale NOS
questionnaire was used by Tsai (1998) for eighth graders. The questionnaire was
Downloaded by [Linköping University Library] at 02:56 04 October 2014

conducted twice in this study, and the Cronbach alpha reliabilities for each survey
were 0.83 and 0.87.

Scoring scheme
All items on the conceptual problem-solving tests were graded using the following
scheme: answers with correct statements and use of scientific concepts were
assigned 3 points; answers with sound arguments and minor mistakes in using the
apparatus or concepts were assigned 2 points; answers with partial misconceptions
but that indicated some degree of relevance toward the scientific concept were
assigned 1 point; and answers with irrelevant statements or misconceptions were
given 0 points. Two high school chemistry teachers who were trained to use this
scoring scheme graded all of the student test answers. The average of the two
teachers’ grading was calculated for each student. The inter-reliability of the
measurements between the two evaluators was 0.95 for the first conceptual
problem-solving test and was 0.96 for the second.
For the post-positivist-oriented items in the NOS questionnaire, a ‘strongly
agree’ response was assigned 5 points and a ‘strongly disagree’ response was
assigned a score of 1. Items representing an empiricist view were scored in reverse.
Therefore, a higher score on this questionnaire represents a stronger belief about
post-positivism.

Participants and procedure

A total of 620 Year 8 students participated in this study. They were enrolled in 16
different classes at four junior high schools near Kaohsiung City in Taiwan. The
students were chosen because of their proximity and availability to the researcher.
Because these students were from families of variety of socio-economic back-
grounds and their academic achievements were varied, it is reasonable to say that
they were typical junior high school students in Southern Taiwan.
In the middle of a school year these students were asked to take the first
conceptual problem-solving test and to respond to the NOS questionnaire. They
then took the second test and the questionnaire again at the end of the school year.
To understand the student problem-solving strategies, four of the students were
selected using the stratified random sampling procedure to participate in follow-up
interviews. Students were categorized as post-positivist-oriented or empiricist-
aligned based on their score on the NOS survey. Students were also assigned as high
 UNDERSTANDING THE NATURE OF SCIENCE AND PROBLEM-SOLVING 105

academic achievers or low academic achievers based on their first mid-term test
result. In both of the classifications, the cut-off point was the median score. These
students were consequently divided into the following four groups: high academic
achievement and high score on the NOS survey; high academic achievement and
low score on the NOS survey; low academic achievement and high score on the
NOS survey; and low academic achievement and low score on the NOS survey.
Because of the exploratory nature of this study and with the belief that students that
had better conceptual understanding would be more expressive in their problem-
solving processes, only the two groups of high academic achievers were selected as
the target group for follow-up interviews. Two students were randomly selected
from each of the two groups. With the use of a sample item, all of the four students
practiced the thinking aloud procedure that was proposed by Larkin and Rainard
Downloaded by [Linköping University Library] at 02:56 04 October 2014

(1984). Following the practice, they were asked to review their answers to the
conceptual problem-solving tests and to carry out the thinking-aloud problem-
solving procedure for one item in each test. This process was audio-tape recorded
and transcribed verbatim. The students were encouraged to speak freely and as
much as they could. The researcher clarified the student’s meaning whenever his/her
speaking was ambiguous.

Quantitative results
Table 1 presents the mean and standard deviations of the subscale scores for the
NOS surveys. The results from the first and the second survey appear to be very
consistent. The mean score in both surveys remained nearly unchanged at 189. The
means of the subscales (i.e. the nature of scientific knowledge, the nature of
scientific method, and the nature of scientific community) in both surveys were also
very close.
The results of the two conceptual problem-solving tests revealed that the
students had relatively low ability in applying scientific concepts. With the first test’s
possible total of 29, the students got a mean score of 8.23 (standard deviation =
4.65). In the second test, the students performed as poorly as on the first. Their
mean score was 4.70 (standard deviation = 3.90) out of the total possible 17. When
the first conceptual problem-solving test score was selected as the dependent

Table 1. Means and standard deviations of the nature of science subscales’

scores.

First survey Second survey

Variable Mean Standard deviation Mean Standard deviation Possible total

Total score 189.91 19.14 189.84 21.34 265

NSKSa 111.47 10.97 112.28 12.48 155
Methodb 35.31 5.51 34.77 5.75 50
Scientistsc 43.12 5.84 42.79 6.09 60

a
The nature of scientific knowledge.
b
The nature of scientific method.
c
The nature of scientists and scientific community.
 106 H.-S. LIN ET AL.

Table 2. Regressional results of the first and the second conceptual

problem-solving tests.

First test Second test

Variable R2 F R2 F

Methoda 0.21 155.47*** 0.22 169.54***

Scientistsb 0.02 22.12*** 0.02 16.96***
NSKSc – – 0.01 4.10*

a
The nature of scientific method.
b
The nature of scientists and scientific community.
Downloaded by [Linköping University Library] at 02:56 04 October 2014

c
The nature of scientific knowledge.
*** p < 0.0001, * p < 0.05.

variable and the subscale scores from the NOS were used as the regressors, the
stepwise multiple regression revealed that the nature of the scientific method alone
explained 21% of the variance. The nature of scientists explained 2% more. Other
independent variables did not explain a significant variance.
To compare this regressional result with the use of a different conceptual
problem-solving test, the same procedure was used for the regression for the second
test. Coincidentally, the nature of the scientific method appeared as the best
predictor for the students’ conceptual problem-solving ability. It explained 22% of
the variance, while the nature of scientists explained 2% more and the nature of
scientific knowledge added another 1% of the variance. All of the regressional
results are presented in table 2.
Independent t-tests were used to examine whether there is any significant
difference in the conceptual problem-solving ability between the post-positivist-
oriented and the empiricist-aligned students. Table 3 shows that the post-positivist
students (mean = 9.96, standard deviation = 4.58) outperformed their counter-
parts (mean = 6.49, standard deviation = 4.03) significantly (p < 0.001) on the

Table 3. t-test result from the first conceptual problem-solving test.

Group N Mean Standard deviation t p

Post-positivist 303 9.96 4.58 9.87 0.000

Empiricist 300 6.49 4.03

Table 4. t-test result from the second conceptual problem-solving test.

Group N Mean Standard deviation t p

Post-positivist 294 6.20 4.07 10.09 0.000

Empiricist 293 3.19 3.06
 UNDERSTANDING THE NATURE OF SCIENCE AND PROBLEM-SOLVING 107

first conceptual problem-solving test. A similar result was obtained when the
statistical analysis t-test was used on the second conceptual problem-solving test.
Table 4 presents the analysis details.

Qualitative results
The preliminary data analysis of the follow-up interviews revealed that the high
achievers with low scores on NOS were more likely to memorize the chemical
formulas and manipulate the numbers blindly in a problem to get an answer
although they were not sure of the meaning of their calculation. Conversely, the high
academic achievers with high scores on the NOS were more likely to look at a
problem qualitatively and retrieve their conceptual understanding to solve the
Downloaded by [Linköping University Library] at 02:56 04 October 2014

problem. The students’ problem-solving strategies are described in the following

paragraphs. This article compared the problem-solving strategy between high (Hank
and Helen) and low (Larry and Laura) NOS scorers of high academic achievers.
These comparisons focused mainly on the strategies for defining a problem and
manipulating data.

Defining a problem
In the density test item, students were asked to identify real gold from two pieces of
metal with identical volume and appearance. The density of gold (19.3 g/cm3 ) and
necessary equipment like graduated cylinders, beakers and a balance were given.
Although all of the four students were high academic achievers, Hank and Helen
successfully solved the problem while Larry and Laura struggled to explain their
strategies. Both Hank and Helen have a thorough understanding of the concepts of
mass, volume, and density, which enabled them to define the problem correctly and
clearly. Despite Hank using the water displacement of the two metals in a cylinder,
while Helen used a beaker instead, they all knew that they had to measure the
volume and mass of the two metals separately to use the density = mass / volume
equation to calculate the exact density of each metal. Larry and Laura had only a
limited understanding of the related concepts. The problem was represented
incorrectly by both of them.
Laura: Put water in the graduated cylinder. Drop the two metals into the cylinder one
at a time and compare the amount of water displaced by each metal. The one
with greater water displacement would be the real gold.
Researcher: Can you explain your procedure with hypothesized data?
Laura: Suppose that the two metals are marked A and B. If A is the real gold, then the
water displacement would be 19.3. If A is measured with the balance, its
weight would be 19.3.

It should be noticed that Laura did not mention the volume and mass units. In
addition, she incorrectly regarded that the real gold’s volume, mass, and density
would all equal 19.3.
Larry: Put the real gold into a beaker (full of water). The mass of the water being
displaced would be the mass of the real gold. If a metal can displace the same
amount of water, then the metal would be real gold.

Larry did not represent the problem correctly. In this item, he was asked to identify
which metal was gold from two similar metals. However, he thought that the real
 108 H.-S. LIN ET AL.

one was known. In addition, he made a mistake by equalizing the mass of the gold
to the mass of the water displaced by the gold.

Data manipulation
In the test item on molecular weight, as shown in figure 2, the students were asked
to calculate the molecular weight ratio of three substances, A, B, and C. The post-
positivist-oriented students were able to use their reasoning skills, based on their
conceptual understanding. For Hank, a thorough understanding of the problem was
critical in order to retrieve the related concepts in his mind. After reading the
problem two times, he explained his solution strategy and performed the calculation
as follows:
Downloaded by [Linköping University Library] at 02:56 04 October 2014

1. Calculate molecular weight of A and B using the atomic weight ratio.

Because the ratio is 3:4 and the A molecule has two atoms, the molecular
weight ratio of A to B should be 6:4.
2. Find the molecular weight of C. Based on the law of mass conservation, mA
+ mB = mC. The molecular weight of C is: 6 + 4 = 10.
3. Therefore, the (molecular weight) ratio of A, B, C is 6:4:10.
Helen, another post-positivist-oriented student, performed a problem-solving
strategy similar to Hank’s. She had a clear mind about the concepts of atom,
molecule, atomic weight, molecular weight and molar ratio. Her problem-solving
procedure in the following explanation revealed the richness of her conceptual
process.
Helen: If one unit volume of A molecules and one unit volume of B can produce one
unit volume of C, then the molar ratio of A, B, C should be 1:1:1. One
molecule of A contains two atoms while the B molecule has only one atom. If
the atomic weight ratio of A to B is 3:4, then, the molecular weight ratio should
be 6:4. The C molecule contains two A atoms and one B atom. Therefore, C
= 2A + B = 6 + 4 = 10. Apparently, the molecular weight ratio is 6:4:10
or 3:2:5.

In contrast to the performance by Hank and Helen, Larry and Laura, the two
empiricist-aligned students tended to hunt for the numbers or the statements that
they felt useful to them without doing any further knowledge retrieval action to
check the appropriateness of their problem-solving actions. When they were asked
the reasons for their strategy, they all explained that the idea came from their
intuition. It can be seen from Larry’s and Laura’s procedure that mindless
calculation does not exhibit any sign of recalling the concept relationship in his/her
knowledge structure.
Larry: Two A atoms react with one B atom and produce C, which has three atoms in
total. Therefore, the molecule weight ratio (of A, B, C) is 2:1:3.
Laura: Suppose x, y, z represent the molecular weight of A, B, C. x / 2 = 2, y / 1 =
1, z / 3 = 3. So x:y:z = 4:1:9
Researcher: Why is that x / 2 = 2?
Laura: Because there are two atoms in the A molecule.
Researcher: How do you know that the molecular weight x dividing by 2 equals to 2?
Laura: I don’t know. Perhaps it’s my intuition.

It should be noted that both Larry and Laura were high academic achievers. They
performed as well as Hank and Helen did in classroom tests, which are focused
 UNDERSTANDING THE NATURE OF SCIENCE AND PROBLEM-SOLVING 109

primarily on a static body of scientific knowledge. However, in a test like those in

figures 1 and 2, which emphasize the dynamic nature and relationship of scientific
knowledge, they performed differently from Hank and Helen.

Discussion and implication in science education

Despite the fact that the follow-up interviews were limited to four students solving
only two test items, this study provides both quantitative and qualitative evidence
that students’ understanding about the NOS is strongly related to their problem-
solving strategies. This result lends support to the argument by the American
Association for the Advancement of Science (1989) that promoting students’
understanding of the NOS is important because better understanding of the NOS
Downloaded by [Linköping University Library] at 02:56 04 October 2014

might result in higher conceptual problem-solving ability. In addition, the

knowledge of what subscale of the NOS is related to the conceptual problem-solving
ability is helpful, since it provides a way of improving that ability. While De Berg
(1989) indicated that integrating the history of science in science teaching is
desirable, knowing how to integrate is critical for those who are interested in this
way of teaching science. Previous studies found that the use of science history
promoted student understanding about the nature of science (Lin and Chen 2002,
Solomon et al. 1992). The result of this study seems to encourage curricular
developers to focus more on the nature of the scientific method. In other words, the
information regarding how and why previous scientists conducted experiments,
discussed, or debated to defend their hypotheses should be emphasized. Although
more studies should be conducted to confirm the practical utility, the emphases
must be on the potential for benefiting students understanding the nature of the
scientific method. This in turn might enhance student problem-solving ability at the
same time.
The ‘scientific method’ subscale includes 10 items (see figure 3) from the
questionnaire. Seven out of the 10 items are empiricist-aligned statements while

Post-positivist-oriented statements
#7 There are various methods in conducting scientific inquiries. The selection of methods depends
on circumstances.
#46 It is not guaranteed that careful and precise scientific method would result in success.
#48 Scientists pre-determined some aspects of investigations while they are conducting
experiments.

Empiricist-aligned statements
#5 Scientific problems can be solved if you follow the single scientific method that is universally
recognized in the scientific community.
#8 When two scientists investigate one same natural phenomena, they are going to present a similar
report.
#11 The explanation of a scientific experiment is fixed. It would not be changed with different
persons.
#17 There is always only one appropriate scientific method for solving a scientific problem.
#21 Scientists follow the same rule to select scientific methods.
#30 Scientists do not know or expect outcomes of their experiments.
#49 Scientific research studies are restricted to follow the inductive or deductive method.

Figure 3. Ten items of the ‘Scientific method’ subscale.

 110 H.-S. LIN ET AL.

three items are post-positivist-oriented statements. Most of the statements are

related to scientists’ expectation of their experiments, the selection of scientific
methods, and the interpretation of science experiments. The initial results of the
study revealed that students are likely to perform better on conceptual problem-
solving if they have better understanding on the following contemporary model of
the NOS:
1. Observations and explanations of scientific experiments are theory-laden.
2. There is no single scientific method. The methods used by scientists are
dependent on circumstances.
3. Scientists are not compelled to use the traditional scientific method.
To investigate the relationship between views of science and problem-solving
Downloaded by [Linköping University Library] at 02:56 04 October 2014

strategy, we examined student performance on the problem-solving tasks and

compared the performance of the two belief groups. It was found that the students
who hold an empiricist-aligned view and tend to memorize information did as well
on school tests that do not require knowledge integration as those who hold a post-
positivist-oriented view and incline to develop integrated understanding. In
contrast, when conceptual understanding is emphasized in the assessments, then
students with a post-positivist-oriented view would be more successful than
students with an empiricist-aligned view. For example, while solving the molecular
weight problem, the post-positivist-oriented group students, who believe that there
is no single scientific method and that the methods used by scientists are dependent
on circumstances, are capable of retrieving related concepts of atom, molecule,
atomic weight, and molecular weight and integrating the knowledge of mathemat-
ical meaning of ratio and the chemical concepts to solve the problem. On the other
hand, the empiricist-aligned students, who believe that the traditional scientific
method is the only guide for inquiry, intuitively grasp the numbers available in the
problem and focus on the formulas they memorized to do meaningless calculations.
These results suggest students’ understanding about NOS plays a significant role in
problem-solving.
The quantitative results from this study revealed that student understanding
about the nature of the scientific method explained 22% of the variance in
conceptual problem-solving ability. The post-positivist-oriented students out-
performed their counterparts in the two conceptual problem-solving tests. Most
prediction studies conducted regression using a single dependent variable once
only. This study selected conceptual problem-solving ability as the dependent
variable and collected the data twice in one year. The comparison of the two
regressional results gave the researchers greater confidence in drawing the
conclusion about the relationship between student understanding about the NOS
and their conceptual problem-solving ability. Future studies can use an experi-
mental design to examine whether the emphasis on the nature of the scientific
method can really result in higher student achievement in problem-solving.
The qualitative result of this study revealed that students with different
scientific epistemological beliefs perform differently in their conceptual problem-
solving strategies. The post-positivist-oriented students were able to apply reasoning
skills in organizing the related concepts retrieved from their knowledge structure.
The empiricist-aligned students tend to grab the numbers and statements in the
problem and perform blind mathematical calculations without checking the
appropriateness of the calculations, and simply ignore the relationship between the
 UNDERSTANDING THE NATURE OF SCIENCE AND PROBLEM-SOLVING 111

related concepts to the problem. Although the mechanism of how and why thorough
understanding of NOS results in better achievement in conceptual problem-solving
needs to be scrutinized, this difference can be partly explained by the findings of
Songer and Linn (1991) and Tsai (1998). In Songer and Linn’s study, it was found
that the students who view science as static assert that science consists of a group of
facts that are best memorized. On the other hand, those who view science as
dynamic believe that scientific ideas develop and change, and that the best way to
learn these ideas is to understand what they mean and how they are related. In Tsai’s
study investigating junior high school student cognitive structure, the students with
better understanding about the NOS tended to recall more information, as well as
show more richness, more flexibility and a higher precision of knowledge recall than
students having an empiricist view. This study further augments and extends the
Downloaded by [Linköping University Library] at 02:56 04 October 2014

fact that students’ understanding of NOS, especially the subscale of scientific

method, is significantly associated with their problem-solving performance. The
results contribute to the growing literature on the investigation of the relationship
between the learning of science and understanding the nature of science.

References
AIKENHEAD, G. and RYAN, A. G. (1991). Students’ views on the epistemology of science (Ottawa: Social
Science and Humanities Research Council of Canada).
American Association for the Advancement of Science (1989). Science for all Americans
(Washington DC: American Association for the Advancement of Science).
BRICKHOUSE, N. (1990). Teachers’ beliefs about the nature of science and their relation to
classroom practice. Journal of Teacher Education, 41, 53–62.
CAREY, R. L. and STAUSS, N. G. (1970). An analysis of experienced science teachers’
understanding of the nature of science. School Science and Mathematics, 70, 366–376.
CONANT, J. B. (1957). Harvard case histories in experimental science (Cambridge, MA: Harvard
University Press).
DE BERG, K. C. (1989). The emergence of quantification in the pressure–volume relationship for
gases: a textbook analysis. Science Education, 73, 115–134.
DUSCHL, R. (1990). Restructuring science education (New York: Teachers College Press).
FRAZER, M. J. (1982). Nyholm lecture: solving chemical problems. Chemical Society Reviews, 11,
171–190.
GABEL, D. and SHERWOOD, R. (1980). Effect of using analogies on chemistry achievement
according to Piagetian level. Science Education, 64, 709–716.
GABEL, D., SHERWOOD, R. and ENOCHS, L. (1984). Problem solving skills of high school chemistry
students. Journal of Research in Science Teaching, 21, 221–233.
GALLAGHER, J. J. (1991). Prospective and practicing secondary school science teachers’ knowledge
and beliefs about the philosophy of science. Science Education, 75, 121–134.
KLOPFER, L. E. and COOLEY, W. W. (1963). The history of science cases for high schools in the
development of student understanding of science and scientists. Journal of Research in
Science Teaching, 1, 33–47.
LARKIN, J. H. and RAINARD, B. (1984). A research methodology for studying how people think.
Journal of Research in Science Teaching, 21, 235–254.
LEDERMAN, N. G. and ZEIDLER, D. L. (1987). Science teachers’ conceptions of the nature of
science: do they really influence teacher behavior? Science Education, 71, 721–734.
LEE, K. W. (1985). Cognitive variables in problem solving in chemistry. Research in Science
Education, 15, 43–50.
LIN, H. (1998). The effectiveness of teaching chemistry through the history of science. Journal of
Chemical Education, 75, 1326–1330.
LIN, H. and CHEN, C. C. (2002). Promoting pre-service chemistry teachers’ understanding about
the nature of science through history. Journal of Research in Science Teaching, 39, 773–792.
LIN, H., SHIAU, B. and LAWRENZ, F. (1996). The effectiveness of teaching science with pictorial
analogies. Research in Science education, 26, 495–511.
 112 UNDERSTANDING THE NATURE OF SCIENCE AND PROBLEM-SOLVING

MACKAY, L. D. (1971). Development of understanding about the nature of science. Journal of

Research in Science Teaching, 8, 57–66.
NAKHLEH, M. B. (1993). Are our students conceptual thinkers or algorithmic problem solvers?
Journal of Chemical Education, 70, 52–55.
NAKHLEH, M. B. and MITCHELL, R. C. (1993). Concept learning versus problem solving, there is
a difference. Journal of Chemical Education, 70, 190–192.
OSBORNE, R. J. and WITTROCK, M. C. (1983). Learning science: a generative process. Science
Education, 67, 489–508.
RUBBA, P. and ANDERSEN, H. (1978). Development of an instrument to assess secondary school
students’ understanding of the nature of scientific knowledge. Science Education, 62,
449–458.
RUBBA, P., HORNER, J. and SMITH, J. M. (1981). A study of two misconceptions about the nature
of science among junior high school students. School Science and Mathematics, 81,
221–226.
SELVARATNAM, M. (1983). Students’ mistakes in problem solving. Education in Chemistry, 20,
Downloaded by [Linköping University Library] at 02:56 04 October 2014

125–132.
SOLOMON, J., DUVEEN, J., SCOT, L. and MCCARTHY, S. (1992). Teaching about the nature of
science through history: action research in the classroom. Journal of Research in Science
Teaching, 29, 409–421.
SONGER, N. B. and LINN, M. C. (1991). How do students’ views of science influence knowledge
integration? Journal of Research in Science Teaching, 28, 761–784.
TSAI, C. C. (1998). An analysis of Taiwanese eighth graders’ science achievement, scientific
epistemological beliefs and cognitive structure outcomes after learning basic atomic theory.
International Journal of Science Education, 20, 413–425.