Development of a Paradigm-Oriented Instrument to Assess

In-service and Preservice Science Teachers’ Understanding

of Traditional View of Nature of Science (TVONS)

Lin, Tsai-Ku

Department of Physics, KaohisungNormalUniversity

TEL: 07-7172930-3263

FAX: 07-7111681

Email: .

Abstract: The purpose of this study was to develop a reliable and valid instrument for use with science teachers that might have a strong belief on Newtonian paradigm. Newtonian paradigm was the basis of positivism and logical empiricism, which have the traditional view of nature of science (NOS) and have been shown to disagree with the view of modern science and philosophy of science. The conceptions of paradigm-oriented NOS, which have educational and social importance, may be a significant influence in the teaching of science as a process of thinking. Thus, concern with teachers’ conceptions of paradigm-oriented NOS and their teaching, which has not been investigated before, served as the justification for this study. Evidence in support of the validity of the instrument constructs was obtained by the results of the validation study, which indicate that this instrument has a satisfactory content, construct and criteria–related validity for the purpose of this study. The computation of Cronbach alpha reliability coefficient (α＝0.91，N＝550) suggested that this instrument has high internal consistency. The results of this study provide both summative evaluation information and diagnostic data for science teacher educators to identify the strengths and weakness of a teacher education program as a process of thinking.

Different Paradigm Provides Different View of NOS

According to Kuhn’s famous book <The structure of scientific revolutions> (1970, 2nd edition), a mature science always has a paradigm, and reasoning works by following the standards of the paradigm. Members of a mature scientific discipline learn key concepts, puzzle-solving techniques and what sorts of thing constitute important puzzles by working through solutions to puzzles described in such works. In the course of their learning, they imbibe deep-seated assumptions about the nature of the world and about what observations and concepts are relevant for properly conducting research in that discipline. Their understanding of key scientific concepts, their perceptions and their style of reasoning will all have been shaped by the paradigm. When paradigms change, scientists see the world of their research-engagement differently.

The Problem

In many countries, the dominated paradigm of science teachers education program was Newtonian paradigm. Newtonian paradigm is an effective way to teach scientific knowledge, but the views of NOS proposed by this paradigm has been modified by modern science (Fields theory, Relativity, Quantum physics or Modern System theories) and philosophy of science (e.g. AAAS, 1989; McComas, 1998 ). Some researches have claimed that the teacher’s image about the NOS can be conveyed to their students explicitly or implicitly, because the teacher’s theory or knowledge about science affects his or her classroom behavior, such as decision-making about texts, curriculum, lesson preparation, assessment and other pedagogic matters. (Matthews, 1994). In Taiwan, due to the background of Chinese culture, teachers play a key role in science teaching. Therefore, it is important to develop a Newtonian paradigm-oriented instrument to assess teachers’ views of the NOS.

The Model Factors of TVNOS

The main views of NOS of Newtonian paradigm can be found in a few lines from the preamble to a draft written in 1703 by Newton of “a scheme for establishing the royal society”:

“Natural philosophy consists in discovering the frame (structure) and operations (exertion of forces) of Nature, and reducing them, as far as may be, to general Rules of Laws, ---establishing these rules by observation and experiments, and then deducing the causes and effects of things.” (ms cit. Westfall 1980, p632)

This scheme indicates, explicitly or implicitly, the following epistemological and methodological aspects of the Newtonian paradigm:

  1. The general goal of natural philosophy (natural sciences) is to discover the structure, the exertion of forces, and the general rules of Nature, which are already out there (an objective existence).
  2. The structure of the objective world is constructed by some invariant reality (corpuscles), and is a static existence inside the absolute space and time (which is the axiom of the Newton paradigm).
  3. The scientific method is used to deduce one thing from another (infer or draw as a logical conclusion, or derive by a process of reasoning). The general rules or laws are established by (or induced from) observations and experiments. (i.e., the solid foundation we can give to science itself must be based on observations, experiments or empirical evidences).
  4. The general rules or laws are universally causal (not necessarily causal themselves, but only that they permit the deducing (in the sense of inferring) of causal relations). Chance, uncertainty, and probability play no role in these exact anddeterministic laws.
  5. The general rules or ‘laws of Nature’ are true for all time.

Based on this scheme, we proposed four dimensions (two for epistemology and two for methodology) and nine model factors for TVNOS instrument. The overall statement is as following (ref. appendix I):

Science (or scientific knowledge, theory, method and activity) is not absolute, objective, invariant, exact or deterministic. Science is not entirely empirical, logical, generally applicable, genuinely reliable or rational.

Instrument Development and Field Testing

The development and field testing of the instrument took place in several steps. These steps are outlined below. Building reliability and validity into the instrument were primary concerns through the process.

Step 1: Establishing a Model of the NOS of Newtonian paradigm.

As shown above. To establish the validity of the model, it was submitted to three physicists for review. The reviewers were asked to independently critique the model in light of the following questions:

  1. Does the model reflect the prominent attributes of Newtonian paradigm as they are generally held among physics?
  2. Does the model constitute an inclusive model of the nature of Newtonian paradigm?
  3. Are the model factors exclusive?

After a review of the model by the physicists and a revision of the model by the investigators, the model was endorsed as valid by each of the physicists.

Step 2: Item Pool Preparation

A five-point Likert-type scale was selected as the instrument type to be developed. Five to seven positive affect item statements and the same number of negative affect item statements were written for each of the nine factors of the model. History and philosophy of science (e.g.Aicken, 1991; Couvalis, 1997; Holton, 1985; Hepel, 1966; Hisenberg, 1958) and the related statements in existing NOS instruments (e.g.Kimball, 1967; Rubba, 1976; Cotham, 1981; Nott, 1995; AAAS, 1989; Cleminson, 1990; Aikenhead, 1991; Lederman, 1990, 1992, 1997, 1998; McComas, 1998; Eflin, 1999) were the main sources of the item statements. Care was taken in writing the item statements to make them science content general, to use simple wording, and to eliminate item statement ambiguity.

Step 3: first item refinement- reading level

The first step in the item refinement process was an attempt to eliminate respondent reading level as a confounding factor by assuring that the item statements were at the science teachers reading level. Five senior science teachers were selected to participate in an examination of the item statements. Based on the majority’s suggestion, some of the item statements have changed and some of them dropped from the pool leaving 100 item statements.

Step 4: Second Item Refinement--Form and Content

The surviving item statements were submitted to 23 physics teachers for evaluation of form and content. They were asked to work independently and to evaluate the item statements with respect to the following questions:

  1. Is each of the positive item statements resonant with its respective model factor?
  2. Is each of the negative item statements consonant with its respective model factor?
  3. Is the content of item statements general in their domain of science reference?

The participants entered revision suggestions and comments directly on each item statement. Thirty-seven item statement pairs survived after revisions were made.

Sept 5: Third Item Refinement –A Tryout

The item statements were each attached to a five-point Likert scale labeled “strongly agree”, “agree”, ”neutral”, “disagree”, “strongly disagree” and randomly arranged as a tryout instrument. The tryout instrument was administered to 89 preservice physics and chemistry teachers. An open discussion of the items was held with the students immediately following administration of the instrument. 55 items (not necessarily item pairs) survived the procedure.

Step 6: Item Selection Panel—Judged Item Content Validity

The content validity of the 55 item statements was judged against the Model of TVNOS by a panel of 9 experts. The panel was composed of three physicists, three science educators, and three graduate students (one doctoral student from the graduate school of science education, two master students from physics department, they are all senior science teachers too).

The panel members worked independently comparing items to the model factor for which each was written. For each item the panel members were asked to consider whether a respondent’s strong agreement with it could be taken as evidence that the respondent:

  1. Assumes the position expressed by the respective model factor.
  2. Assumes the position opposite to that expressed by the respective model factor.
  3. Assumes both the position expressed by the model factor and the position opposite to that expressed by the respective model factor.
  4. Assumes neither position

Marginal comments were also solicited from the panel members on item statement revisions they felt were necessary.

Twenty-five positive and twenty-five negative affect item statements were judged by at least seven of the panel members to measure the respective factor in the model of the nature of TVNOS after two rounds of item statement judging and revision.

Step 7: Field Testing and Item Selection

The fifty items which were judged content valid by the panel of experts were each attached to the five-point Likert scale previously used and randomly arranged as a tryout instruments. The tryout instrument was administered to the 550 inservice and preservice science teachers enrolled in five major (physics, chemistry, biology, industry education, and science education) at Kaohsiung and Jang-Huah Normal university in Taiwan.

The discrimination quality of the items was identified by calculating a Pearson product-moment correlation coefficient, r, between each item and the total score for the 12 items corresponding to the respective model factors (negative items were scored in reverse). 100 such correlation coefficients were calculated; one for each item. The correlations ranged from r=0.08 to r=0.82.

The Instrument

The 50 most discriminating-reliable combination of items were randomly arranged to form the TVNOS. Subscales were composed from the 14 or 16 items, half negative and half positive, corresponding to each of the four dimensions and 5 to 7 items corresponding to each of the ninemodel factors. Table 1 presents the item-to-subscale key. The TVNOS items are displayed in Appendix II.

In scoring the TVNOS, the point value from 1 to 5 was assigned for each item. Subscale scores were calculated by adding the appropriate positive and negative items, and the total score of TVNOS was obtained by summing these subscale scores. The total score and all the subscale scores were normalized to 100 points maximum.

TVNOS Reliability

The reliability of the TVNOS was assessed during its development and in subsequent studies with samples of In-service and preservice science teachers. A brief description of the samples and the respective coefficient alphas are listed in Table 2.The computation of Cronbach alpha reliability coefficient, α=0.91 for 19 different groups and 511 effective samples, suggested that this instrument has high internal consistency.

Test-retest reliability was also established on the TVNOS with four groups, 43 senior physics teachers, 22graduate students (also senior physics teachers), 37 junior and 12 senior physics students. 13 weeks spaced the test administrations. The Pearson product-moment correlation coefficients, calculated between the test and retest were r=0.70 (p=0,033*, N=12), r=0.60 (p=0000**, N=37), r=0.34 (p=0.031*, N=43), and r=0.42 (p=0.002**, N=22) respectively.

TVNOS Construct Validity

The construct validity of TVNOS was examined after its development by testing an anticipated difference in understanding the NOSbetween two groups of science teachers owing to differences in their instructional backgrounds. Science teachers completing an history of science course were expected to understand the NOS better than teachers who have not experienced the course. The mean scores of the two groups on the TVNOS and its subscales, before and after they took the history of science course, were compared using a t-tests technique for independent samples. These means, the respective standard deviations, and the results of the t tests are presented in Table 3 and Table 4.

Physics and chemistry teachers who had studied history of science have higher mean scores on the entire TVNOS and some subscales of TVNOS than before they took the history of science course.

TVNOS Criteria-Related Validity

The criteria-related validity of the TVNOS was examined after its development by bivariate correlation among the pretest and posttest of TVNOS, VOSTS (Aikenhead, 1989) and UNOS (Lin, 1996). It appears that there was no significant correlation among the pretests of TVNOS, VOSTSand NUOS, and significant correlation among the posttests of these three instruments with respect to a paradigm-oriented history of scientific thought (ref. Table 5 and next paper).

Acknowledgement

This paper was supported by a grant from the Taiwan National Science Foundation, 89-2511-S-017-010. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the Taiwan National Science Foundation.

Appendix I: TVNOS model of Science

Overall
/ Science (or scientific knowledge, theory, method and activity) is not absolute, exclusive, objective, invariant, exact or deterministic. Science is not entirely empirical, logical, generally applicable, genuinely reliable or rational.
Absolute / Scientific knowledge, theory, method and activity are not ultimate.
Science’s ongoing process leads to an increasingly better understanding of how things work in the world but not to absolute truth.
Empirical / Induction is not the only way to do science. There simply is no fixed set of steps that scientists always follow, no one path that leads them unerringly to scientific knowledge. The conceptual schemes elaborated by scientists are fundamental to further inventions and the hypotheses that will lead to further inventions.
Logical / Science relies heavily, but not entirely, on logic and mathematics.
Objective / New science is produced by creative acts of the imagination allied with the methods of scientific inquiry. As such science is a personal and immensely human activity (Science is artificial and does not show nature as it really is). Scientists study a world of which they are a part, not a world from which they are apart. For example, scientific evidence can be biased in how the data are interpreted, in the recording or reporting of the data, or even in the choice of what data to consider in the first place. Scientists’ nationality, sex, ethnic origin, age, political convictions, and so on may incline them to look for or emphasize one or another kind of evidence or interpretation.
Generally applicable / Science has a range (limitations) and variety of methods. The particular methods selected on any one occasion depend on the particular circumstances. There is no one method of science applicable at all times. The law of causation is not universal.
Genuinely reliable / Science is fallible. Evidence for the value of this approach is given by the improving ability of scientists to offer reliable explanations and make accurate predictions.
Rational / There are no rational and defensible criteria between science and non-science, rational and irrational, meaning and meaningless, science and meta-science. Science possesses the attribute of openness, an openness of mind, allowing for willingness to change opinion in the face of evidence, and an openness with respect to the investigation, unlimited by such factors as religion, politics, philosophy and arts etc.. Progress in all fields of science depends on intelligence, hard work, imaginations and even chance.
Invariant / Science is dynamic, changing, and tentative. Science does not consist in a static collection of facts. Science is an ongoing process, some times punctuated by striking changes in beliefs and methods. As a result, we can not take current scientific knowledge to be complete and final.
Exact and deterministic / Probability and uncertainty are characteristic of all science.

Appendix II: TVNOS Item Statements

  1. All science must (or should) conform to the rules of logic automatically.
  2. The goal of scientific inquiry is to discover the generally applicable truths of nature.
  3. Progress of science does not need the action of irrational factors, like imagination, conjecture, tuition or speculation.
  4. Progress in science does not come from the static collection of absolute truth.
  5. All events have a cause, but may not be explained by science.
  6. All scientific knowledge is changeable, even they are correct so far.
  7. Scientists can not guarantee that the knowledge they obtain from scientific inquiry is genuinely reliable.
  8. The main goal of scientists is to discover the invariant truth about nature.
  9. Natural law is objective, which is independent of the existence of human being.
  10. Progress in science consists of discovering theories that represent a closer approximation of the absolute truth.
  11. In science, if we apply logic to reason, then we can be sure that we will not get contradictory results.
  12. There is no generally applicable rules of logic that can help us deduce what is undoubtedly true in science.
  13. What is true is logic may not be true in science.
  14. All truth is subject to change, and scientific truth is no exception.
  15. There is no generally applicable theory for doing science.
  16. The more content of logic, the more persuade of the scientific theory, the more likely be accepted by others.
  17. Scientists can develop their theories without empirical evidence.
  18. Scientists can test their theories without the help of empirical evidence.
  19. Scientific theory is not necessarily based on objective observations.
  20. We can use the untested theory to do the scientific inquiry.
  21. There are no logical criteria in science that can be used to determine whether a scientific theory is correct or not.
  22. Only the objective (unprejudiced) observations and experimental results can be used to determine whether scientific knowledge is true or not.*
  23. Scientific theory has functions of explanation and exact predictions.
  24. Scientific theory is stable and invariant after its verification.
  25. If the scientific theory has been proven to be correct, then it can be used to obtain genuinely reliable knowledge.
  26. Empirical evidences (observations, experiments) are the foundation of scientific theory, so the model of scientific development must be “experiments come first, theory follows”.
  27. The scientific methodsaregenuinely reliable, because they are based on accurate observations, measurements and experiments.
  28. Effective scientific methods last forever after they are discovered.
  29. To be sure of approaching nearer to the absolute truth, one should follow the appropriate scientific method.
  30. The scientific method is not just the method which is used to deduce natural laws from empirical evidence（e.g. observations, measurements and experiments）.
  31. One of the characteristic of the scientific method is that it is objective, so the philosophic method which is subjective and intrusive can not become a scientific method.
  32. Scientific inquiry begins with and proceeds through the systematic organization and classification of facts, followed by other logical rules of induction to formulate the laws.
  33. We should try to find out suitable scientific method during the scientific inquiry, not apply generally applicable scientific method to do science.
  34. Even we have applied the correct scientific method, we may not obtain the genuinely reliable results.
  35. The observations made and experiments conducted by well-trained scientists can be influenced by their personal feeling, theory or belief.
  36. There is no scientific method can guarantee that it can obtain the genuinely reliable truth.
  37. The observations made and experiments conducted by well-trained scientists may not be accurate.
  38. There exists an effective scientific method that can be used to get genuinely reliable results.
  39. Scientific activityis not necessarily a rational activity; itis shaped by the personal beliefs and political attitudes of its practitioners and reflects, in part, the history, power structure, and political climate of the supportive community.
  40. The formations of scientific hypotheses involve irrational factors, which can not be explained by logic or experiments.
  41. Scientists need not apply the generally applicable scientific method and scientific activity to conduct scientific inquiry.
  42. Science is a rational enterprise; scientific activities are rational activities too.
  43. What we know about the moon (or Nature) is the creative and successful conjectures (imagination) of scientists, not necessarily objective truth about the moon (or Nature)
  44. There is a general logical procedure in science, which can be applied to formulate inductively scientific hypotheses from sets of specific observations and experimental data.
  45. If human civilization was restarted from the beginning, I think it would end up in the same situation as we have now.
  46. Science is not propelled exclusively by its own internal logic; it is shaped by the personal beliefs and political attitudes of its practitioners etc.
  47. Scientific activity is the processes of the absolute truth about nature.
  48. Wedon’t have to use verified scientific hypotheses as the exclusive basis for scientific explanations.
  49. The processes of scientific activity follow the objective law, there is no place for conjectures or lucks.
  50. Scientists and scientific activitiesare rational, they do not affect by social or personal psychological factors.

TVNOS Item-to-Subscale Key（★=Negative Item）