區塊研究---國小中高年級科學本質能力指標之課程、教學與學習的評量工具之研究

行政院國家科學委員會專題研究計畫 成果報告

區塊研究--國小中高年級科學本質能力指標之課程、教學 與學習的評量工具之研究(第 3 年)

研究成果報告(完整版)

計 畫 類 別 ： 整合型

計 畫 編 號 ： NSC 95-2522-S-011-001-MY3

執 行 期 間 ： 97 年 08 月 01 日至 98 年 07 月 31 日 執 行 單 位 ： 國立臺灣科技大學技術與職業教育研究所

計 畫 主 持 人 ： 陳素芬

計畫參與人員： 碩士級-專任助理人員：李偉豪

碩士班研究生-兼任助理人員：蘇暐珍 大專生-兼任助理人員：張名棻

處 理 方 式 ： 本計畫涉及專利或其他智慧財產權，2 年後可公開查詢

中 華 民 國 98 年 10 月 29 日

行政院國家科學委員會補助專題研究計畫 5 成 果 報 告

□期中進度報告

區塊研究-國小中高年級科學本質能力指標之課程、教學與學 習的評量工具之研究

計畫類別：□個別型計畫 5 整合型計畫 計畫編號：NSC 95－2522－S－011－001－MY3

執行期間： 95 年 8 月 1 日至 98 年 7 月 31 日

計畫主持人：陳素芬 共同主持人：

計畫參與人員：蘇暐珍、李偉豪、張名棻、黃秀莉、洪志嘉、張素芬、

蔡本慧

成果報告類型(依經費核定清單規定繳交)：□精簡報告 5完整報告

本成果報告包括以下應繳交之附件：

□赴國外出差或研習心得報告一份

□赴大陸地區出差或研習心得報告一份

□出席國際學術會議心得報告及發表之論文各一份

□國際合作研究計畫國外研究報告書一份

處理方式：除產學合作研究計畫、提升產業技術及人才培育研究計畫、

列管計畫及下列情形者外，得立即公開查詢

□涉及專利或其他智慧財產權，□一年□二年後可公開查詢

執行單位：國立台灣科技大學

中 華 民 國 98 年 10 月 30 日

中文摘要

本計畫的目的是發展評量工具以評量科學本質課程、國小教師之科學本質概念和教學 信念，以及國小學童對科學本質的認識。評量工具包括學生問卷、教師問卷、課程嵌入式 評量、課堂觀察檢核表，以及教師晤談問題。問卷的發展模式是採 empirical-based approach，意即先以質性方法收集資料，再據以發展問卷題項，試測、訪談、修改與更多 試測、修改，直到可接受的信效度為止。課程教材「溶解」單元之嵌入評量題項則由教師 之行動研究粹取出來，經過兩次與國小教師的討論與修改，再同問卷和晤談的資料比較以 建立信效度。課堂觀察檢核表則是採 Reformed Teacher Observation Protocol 加上 12 個有關 科學本質的題項。教師晤談的問題乃參考 VNOS 和合作教師的教案，以瞭解其科學本質關與 融入教學的情形，並診斷教師的科學本質概念和對於教授科學本質的知能與態度。

關鍵詞：科學本質、國小自然、評量

Abstract

The purpose of this project was to develop valid and reliable instruments for evaluating curricula of the nature of science (NOS), elementary teachers’ conceptions of NOS and attitudes toward teaching NOS, and elementary students’ views on NOS. A student questionnaire, a teacher questionnaire, an embedded assessment for the dissolving unit, an observation checklist, and an interview protocol were developed. The student questionnaire was based on empirical data derived from students’ written paragraphs and teachers’ action research. The draft was field tested, followed by interviews, and then modified to establish its validity and reliability. The teacher questionnaire was an improvement of VOSE (Chen, 2006) in terms of contextualizing the test items. It was used to monitor teachers’ changes in the professional development.

Moreover, the embedded assessment was developed to investigate students’ understanding of the NOS concepts revealed in the dissolving unit. It was validated by the student survey and

interviews. A teacher interview protocol has been used to follow up VNOS and clarify teachers’

NOS views and attitudes toward teaching NOS. Finally, Reformed Teacher Observation

Protocol and extra items related to NOS teaching help to analyze teachers’ teachings and provide feedbacks for improvement. Researchers and instructors may use the instruments to diagnose pre-/in-service elementary teachers’ views on NOS, attitudes toward teaching NOS, and students’

understanding of NOS.

Keywords: nature of science, elementary school science, instrument, evaluation

報告內容

Our Block Study aims to develop curricula that may improve students’ 3C, i.e., competency,

cooperation, and confidence, which are often found to be unsatisfactory in our students.

According to international curricular documents and our curriculum guideline, an understanding of the nature of science (NOS) is essential to this competency and important in students’

scientific inquiry and construction of knowledge (American Association for the Advancement of Science, 1993; National Research Council, 1996; McComas & Olson, 1998; Ministry of

Education, 2003; Turkish Ministry of National Education, 2005). It is one of the eight core objectives in the Science and Technology subject in Taiwan. In the Grades 1-9 Curriculum Guidelines (Ministry of Education, 2003), two or three benchmarks bound to NOS are specified for each level.

However, previous studies have revealed that high school and college students possess naïve views about science (e.g., Abd-El-Khalick & BouJaoude, 1997; Dogan & Abd-El-Khalick, 2008;

Lederman, 1992; McComas, 1996). Clough (2006) concluded that teaching NOS is mostly about conceptual change. However, learners often pick up, or even modify, ideas and

information that fit into their knowledge base, and ignore contrary information (Abd-El-Khalick

& Lederman, 2000; Olson, Clough, & Vanderlinden, 2007; Tao, 2003). Even students who have gained an informed view during interventions may revert easily to their original naïve views, which seem to be more secure to them (Khishfe, 2008; Leach, Hind & Ryder, 2003). It is difficult to change students’ NOS concepts.

Moreover, a few studies have found that elementary students have developed views on NOS and gender stereotypes of scientists (Elder, 2002; Kang, Scharmann, & Noh, 2005; Newton &

Newton, 1998). Kang et al. (2005), Clough and Olson (2004) and Lederman and O’Malley (1990) suggest that NOS be incorporated into science instruction in elementary schools to avoid forming distorted images of science. It is more effective to help students construct informed views early in elementary school, rather than to change their concepts later on. Furthermore, Smith, Maclin, Houghton and Hennessey (2000) revealed that “elementary schoolchildren are more ready to formulate sophisticated epistemological views than many have thought” (p.350).

Elementary school students are capable of and interested in thinking about low-level philosophical questions (Matthews, 1998). Even first graders are able to develop an

understanding of the inferential, tentative, and creative NOS from explicit decontextualized and contextualized instruction (Akerson & Volrich, 2006). In other words, well-designed curriculum and instruction in elementary school could lay a robust foundation for students regarding NOS views. In line with this thought, a set of valid and reliable assessment tools for young students, curricula and elementary teachers would significantly contribute to NOS research. As a result, the purpose of this project was to develop valid and reliable instruments for evaluating NOS curricula, elementary teachers’ conceptions of NOS and attitudes toward teaching NOS, and elementary students’ views on NOS.

研究目的

The purpose of this project was to develop valid and reliable instruments for evaluating NOS curricula, elementary teachers’ conceptions of NOS and attitudes toward teaching NOS, and elementary students’ views on NOS. A student questionnaire, a teacher questionnaire, an embedded assessment for the dissolving unit, an observation checklist, and a teacher interview protocol were developed. These tools provided feedback for the NOS integrated project specifically regarding the development of the NOS curriculum and professional development.

文獻探討

The literature review will focus on the scope of NOS involved in major documents in the field, the empirically-based approach of developing questionnaires, and research on embedded assessment. NOS involves a wide horizon of thoughts in philosophy, history and sociology.

Some of the NOS constructs are based on armchair thinking and are fervently debated among different philosophical schools. Science educators seek to incorporate less controversial NOS issues into science teaching. Previous studies using interviews and open-ended questionnaires such as Views of Nature of Science (VNOS) (Lederman, Abd-El-Khalick, Bell, & Schwartz, 2002) suggest several constructs of NOS for NOS research. However, the constructs are seldom

verified quantitatively by empirical data. Moreover, the interrelation among the constructs is mostly theorizing by researchers, rather than emerging from data. The present study sought to explore a model to holistically explain elementary school students’ views on NOS, which would then provide insights for curriculum development and professional development.

We set out to develop the model by considering NOS issues that are closely related to students’ experiences of school science and are appropriate learning goals for elementary school students. According to international curriculum documents and the literature (e.g., Good, Lederman, Gess-Newsome, McComas & Cummins, 2000; Kourany, 1998; McComas & Olson, 1998; Olson, 2008; Osborne, Collins, Ratcliffe, Millar, & Duschl, 2003; Schwartz & Lederman, 2002; Smith & Scharmann, 1999), the following NOS aspects were taken into account as appropriate to grades 3-6:

1. Nature of scientific knowledge: Scientific claims are based upon reason and empirical evidence. Claims are subject to tests, and experiments or observations should be repeatable by researchers in a variety of settings. Scientific knowledge is durable, yet tentative. It may be modified or dramatically changed in light of new evidence or perspectives.

Innovation of methods and technology and different worldviews all contribute to the change.

Science is done by human beings who are constrained by their society and culture.

Moreover, the development of scientific knowledge is affected by social structures, including questions that are taken to be most significant, and answers that are proposed and accepted (Kitcher, 1998). Scientific knowledge is one of the many forms of knowledge. Other forms of knowledge such as aesthetics and religion are as valuable as science. Science has its limitations and does not solve all problems.

2. Nature of scientific inquiry: Many ways exist to solve a problem. There is no standard

scientific method. Scientists utilize various methods to do science. Diversity of thinking

promotes science development. Specifically, creativity and imagination play an important

role in generating research questions, designing projects and interpreting data. Moreover, scientists conduct research with preconceptions and are affected by their background. They often think before acting and preconceive of what to observe. They attempt to make

objective and accurate observations. Nevertheless, their observations are often theory-laden.

3. Scientific enterprise: Science to some extent reflects social values and viewpoints, and is influenced by the culture in which it is embedded. Science is a creative human endeavor.

Scientific knowledge is not kept for individual scientists, but is published, reviewed, shared and challenged by the community. Diverse individuals, including women and men of various social and ethnic backgrounds, engage in and make contributions to science.

Science is for all.

The curriculum guideline of Taiwan incorporates the nature of scientific knowledge and inquiry. However, it focuses less on scientific enterprise. Social-embeddedness, science community and the engagement and contribution of diverse individuals are all omitted.

Nevertheless, the importance of these issues was recognized in this study. Regarding the engagement and contribution of diverse individuals, the gender issue has been a concern in science education across countries and cultures. Consequently, the gender issue was included in this investigation. Considering the length of the text, this article reports the results, but does not draw on the massive literature regarding gender engagement in science to address the issue in much detail. Rather, the discussion focuses on the rest of the issues that are typically the

concern of research in the NOS field. Table 1 summarizes the NOS issues that were involved in our questionnaires for elementary school teachers and students.

Table 1.國小教師 NOS 與學生 NOS 評量範圍

Elementary school teachers

Elementary students Scientific knowledge

Durable X X

Tentative—graduate change, evolution, revolution X X

Repeatable (Replicable) X X

Testable X X

Consistency X X

logic X

predictable X

Empirical evidence X X

Scientific inquiry

Creative X

imagination X

Scientific method X

Diversity of scientific thinking X X

Limitation of science X X

Observation—objective X X

Accuracy X X

Theory-laden

(hypothesis and prediction)

X X

Scientific enterprise culture

gender X X

race STS

Although an inventory of NOS concepts could be made, it should be noted that neither NOS nor students’ belief systems are fragmentary. However, it is not unusual in research that NOS is collapsed into fragments, and the teaching of NOS is simplified as discussion of several tenets.

In this study, a constant endeavor, including the development of an instrument to obtain empirical data and careful application of statistical methods to analyze the factor structure of the data, was made to realize how various conceptions are interrelated to form a picture of NOS among young learners.

Regarding instrument development, Aikenhead and Ryan (1992) developed an empirically-based approach to generate an instrument entitled the Views on

Science-Technology-Society (VOSTS) for high school students. Their approach emphasized that questionnaire items were written from learner’s perspectives instead of expert’s speculation of possible views. Students’ written paragraphs were analyzed and categorized. Based on the written responses, multiple-choice items were formed and validated twice by written paragraphs and semi-structured interviews. The multiple-choice responses reveal both a student’s position on a question and the reason for choosing the position. Items developed based on learners’

viewpoints could considerably reduce the level of misinterpretation (Aikenhead, 1988). Kang, Scharmann and Noh (2005) and Chen (2006) confirm that empirically-based items yield reliable results. Furthermore, this approach may improve the readability and accuracy of a questionnaire for elementary students to a remarkable extent because language and wording is a particular issue for younger learners. Therefore, this approach was adopted in the present study. The

questionnaire data were then analyzed using up-to-date statistical methods to explore the

constructs of the students’ NOS views. The questionnaire for teachers was revised from Views on Science and Education (VOSE) (Chen, 2006) which followed the same approach.

This project also developed an embedded assessment (EA) for the teaching material, the dissolving unit. EA is characterized as an approach that assessment of students’ learning progress is integrated into instructional materials and is carried out in regular teaching activities (Wilson & Sloane, 2000). EA has been found to be a reliable alternative assessment (Green, Middleton & Reid, 2000). Some educators perceive it as more useful than traditional

achievement/aptitude tests and interest inventories for making educational decisions (Swisher &

Green, 1998; Swisher, Green & Tollefson, 1999). EA can be adjusted to evaluate the quality of

student performance in meeting objectives of reformed curriculum that are often considered

vague and difficult to be measured (Duschl & Gitomer, 1997; Gerretson & Golson, 2005). The

curriculum in question incorporates an EA so that what is assessed is consistent with what is

taught. In other words, students are assessed in the specific instructional context. The EA is

designed according to four principles outlined by Wilson and Sloane (2000). Firstly, it is planned from a developmental perspective, in which progress variables are defined based on students’ prior knowledge and curricular objectives. In our case, we analyzed students’ concepts in several NOS aspects from the literature and interviews with elementary teachers. Moreover, we unpacked the NOS-CIs delineated in the national curricular guideline. Students’ NOS concepts and the unpacked indicators helped us to determine a set of progress variables to structure the instructional and assessment tasks.

Secondly, there must be a match between instruction and assessment. Assessment is not for its own sake; rather it is well-embedded in the instructional activities. Therefore, the instructional materials and assessment tasks were developed simultaneously. Both are continuously adjusted to create a curriculum which emphasizes explicit teaching of NOS and also reveals students’

understanding of NOS.

The third principle is concerned with teacher management and responsibility. “Any

successful classroom-based system must take into account the demands placed on the teacher for administering, scoring, interpreting, and reporting student performance (Wilson & Sloane, 2000, p. 191).” We intended to make the assessment as manageable to teachers as possible by

simplifying the scoring task and integrating the assessment into a well-structured instructional model. On the one hand, the number of open-ended items was reduced to the minimum. Possible response of the assessment items embedded in the learning sheet is listed in the teachers’ manual.

On the other hand, the items were integrated with the teaching content and were instructionally meaningful so that teachers might not feel awkwardly in carrying out the assessment (see Figure 1).

Finally, the fourth principle, quality evidence, addresses the validity, reliability, and fairness,

which are required for all assessment tools. Multiple methods, including observation protocols,

achievement tests, instructors’ ranking, interviews, and surveys, could be utilized to ensure the

quality of the embedded assessment. In this project, students’ views on NOS are collected from a

questionnaire in addition to the embedded assessment. The items on the questionnaire are still

under revision and will be better contextualized in a story. Use of add-on assessment helps to

monitor the quality of the embedded assessment.

Figure 1. Assessment items embedded in worksheet.

研究方法

Students’ Views on Nature of Science (SVNOS).

An instrument entitled Students’ Views on Nature of Science (SVNOS) was developed to investigate elementary students’ conceptual structure of NOS, using students’ ideas about science.

Four major stages were used to establish its reliability and validity. First, the items were drafted according to written responses, structured interviews and excerpts from books. Second, the draft was tested. Then, an item analysis was conducted to refine the questionnaire. Finally, the questionnaire was validated by a second test. A model was derived to explain the NOS views of elementary school students based on the results of SVNOS.

Draft. In the first stage, the items were drafted based on empirical data from elementary

students’ actual ideas about science. Firstly, 431 sixth graders responded to four open-ended questions: What is science? What is needed for scientific knowledge to be established? Do you believe the scientific knowledge in the textbook and Why? Do you think that science can solve every problem and Why? Their complete statements were collected, analyzed and classified into different NOS subcategories by the primary investigator and a research assistant. Secondly, we surveyed available articles, theses and books in the NOS field to look for quotations from classroom dialogues (e.g., Chiu & Kao, 2004, 2006; Wang, 2006). The dialogues were sorted into the same categories as the grade six students’ ideas.

However, the written responses and theses provided inadequate statements about

observations and the gender issue. Students seemed not to relate the open-ended questions to observations. Nor is it common for elementary schools to integrate the nature of observation into their teachings. Moreover, we were interested in whether students have gender stereotypes.

Thus 12 sixth graders were interviewed, focusing on the observation and gender aspects. Many students responded that observation means filling out the data table on the worksheet. They also stated that the purpose of observation is to compare differences between two cases or changes over a period of time. A few of them mentioned that we have some ideas about what to observe before making observations, and oftentimes we want to prove our ideas when making

observations. Regarding gender issues, they felt that scientists are mostly male. Some of them attributed the discrepancy to male’s thinking style and logic. Others attributed it to external causes. For example, one girl stated that although men and women can both engage in science, men are more likely to be successful in science because they, unlike women, who are often

occupied by trivial work such as housework, can concentrate on scientific research. This gender issue is indubitably important if the goal of science for all is to be achieved. Nevertheless, limited by the length of the text, while the results are presented, an in-depth discussion of this issue is omitted. The following discussion focuses on the NOS aspects which are generally covered in the NOS field.

Finally, 83 statements were selected and slightly revised to form the items of the draft

SVNOS. Thirteen NOS aspects were identified: durability, tentativeness, cumulative view,

repeatability, consistency, based on empirical evidence, publish, socio- and

cultural-embeddedness, creativity and imagination, scientific method, the limitations of science, observation (being objective, accurate and theory-laden), and science for all (non-gender-biased).

There were at least 5 items for each aspect. Twelve leading questions were generated to organize the items, for which the 83 statements were possible responses. Each question was followed by a few items, such as:

1. What is “observation”?

1-1. Before making observations, people already have ideas about what to observe.

1-2. Before making observations, people first prepare a table and then fill it out as they make their observations.

1-3. When making observations, people do not need to focus on specific things. They only need to write down what they see.

1-4. The purpose of observation is to compare different cases or changes over a period of time.

1-5. Everyone has his/her own ideas and wants to prove them when making observations.

The items were Likert-type with response options represented by sad faces, smiling faces and question marks to symbolize degree of dis-/agreement and do not understand. The face validity and coding category were verified by a team of five science educators in the NOS field. The questionnaire as well as the corresponding NOS aspect for each item was emailed to the team.

The experts commented on the wording of the items, the coding category to which an item belonged under the context of the guiding question, and the aspects of NOS covered.

Disparities were solved through a teleconference, during which students’ interview and written responses were read out by the primary investigator as needed.

Test 1. The questionnaire was tested on 1,139 third to sixth graders (50.7% boys, 49.3% girls)

at six schools, including four rural schools and two urban schools. In Taiwan, students of urban schools normally have better academic achievement, more parental involvement and higher socioeconomic status. The class size varied with the popularity of the school. Three of the participating schools had class sizes of between 29 and 34. The other three schools had 14-25 students per class. They were invited to complete the questionnaire during the last week of the school year. Most of the students completed the questionnaire in 30 minutes. Twenty-four students from 3

to 5

grades were interviewed following the survey. They were asked to explain each question and item and why they agreed/disagreed with it to provide insights to the questionnaire results.

Item Analysis. Item analysis, exploratory factor analysis (EFA) and structural equation

modeling (SEM) were performed to delete inappropriate items, to investigate possible factor

structures and to identify the best fit structure. The five text anchors from strongly disagree to

strongly agree were coded as 1-5. Do not understand and did not respond were coded as 99 and

0 respectively. During the item analysis, we paid particular attention to the percentage of these

two responses for each item. They are often treated as missing data in social science research,

but missing data may bias the results. Moreover, research has found that with 10% of the data

randomly missing, 59% of the data could be lost (Kim & Curry, 1977), and 35% of the statistical

power could drop (Quinten & Raaijmakers, 1999) using leastwise deletion, the default method for

statistical analysis in most software packages. Therefore, items with higher than 10% of data

missing were eliminated, and those with relatively high but less than 10% of data missing were critically evaluated and revised.

In further factor and statistical analyses, these data were replaced with values estimated by the Expectation-Maximization algorithm in SPSS Missing Value Analysis. The FACTOR program (Lorenzo-Seva & Ferrando, 2006) was applied to explore factors. The number of factors was determined by the parallel analysis (Hayton, Allen, & Scarpello, 2004). Unweighted least squares and promax rotation were used to investigate the constructs of the scale.

Second-order construct was probed by Schmid-Leiman analysis (Schmid & Leiman, 1957).

Previous studies have shown that Schmid-Leiman solution is superior to other multilevel factor analyses such as oblique factor modeling and higher-order confirmatory factor analysis (Gignac, 2007; Wolff & Preising, 2005). Schmid-Leiman solution estimates direct loadings of first and higher order factors on observed variables. Independent contributions of higher order factors and first order factors to variables can be calculated from the factor loadings. The results provide information on whether accuracy, represented by the first order factors, or generality, represented by the higher order factors, should be retained to interpret the data.

According to the item analysis, 47 items organized by seven questions were selected for SVNOS. EFA showed that these 47 items measured seven factors, namely: theory-ladenness (F1), creativity and imagination (F2), tentativeness (F3), durability (F4), coherence and

objectivity (F5), science for girls stereotype (F6), and science for boys stereotype (F7). These factors will be elaborated in the results section. The EFA solution was verified by confirmatory factor analysis in LISERL using two independent data sets. The first set, upper grades data, consisted of 575 5

and 6

grade students. The second set, middle grades data, consisted of 564 3

and 4

grade students. Finally, the average scores of items in each factor were used for descriptive and inferential statistics. The preliminary results will be discussed shortly.

Test 2. The 47-item SVNOS was validated by a sample of 1,091 students (50.6% boys,

49.4% girls), 565 in the 5

grade and 526 in the 6

grade, from 10 schools where individual teachers were recruited for a larger NOS project. None of the students had received explicit NOS instruction. Seven of the schools were located in rural areas and three in cities. 308 participants in two schools, 406 participants in six schools, 192 participants in one school and 185 participants in another school were from the east, west, south and north of Taiwan, respectively. The Test 2 data were also submitted to EFA and SEM analyses.

Views on Science and Education 3 (VOSE-3)

VOSE-3 was an improvement of VOSE (Chen, 2006). Like VOSE, it measures both teachers＇ conceptions of NOS and their attitudes toward teaching NOS. However, the items no longer retain a yes/no position so to reduce the confusion caused by the wording.

Furthermore, we compared a contextualized version and a non-contextualized version with 274 college students followed by 28 students. The results showed that the former has better validity.

The sample items are such as:

題組 2.後來甲和乙各提出一套不同的理論，都可解釋動植物的基因轉植，你認 為科學家們會如何判斷和選擇這兩個理論呢？

非 常 不 同

不 同

沒 意 同

非 常 同

意 意 見 意 意

2-1.當科學家們還不能客觀區分兩套理論的優劣時，會暫時都接受。 0 1 2 3 4

2-2.可能只是理論解釋的方向不同，沒有優劣之分。 0 1 2 3 4

2-3.科學家們通常會接受他們比較熟悉的理論。 0 1 2 3 4

2-4.科學家們會選擇接受比較簡單明瞭的理論。 0 1 2 3 4

2-5.科學家們比較會選擇學術地位較高的人所提的理論。 0 1 2 3 4

2-6.不違背現存重要科學理論的新理論比較會被科學家們採用。 0 1 2 3 4

2-7.科學家們會以直覺判斷。 0 1 2 3 4

2-8.科學家們不會在以實驗證據分出高下之前先接受其中一個。 0 1 2 3 4

Embedded assessment

Assessment items were developed based on Wilson and Sloane’s (2000) four principles mentioned above. For the dissolving unit, there are two or three items in or followed each

teaching activity. Table 2 lists the targeted NOS-CI, the unpacked concepts of NOS-CI, and the corresponding assessment in one of the activities. An example of worksheet which presents two assessment items can be found in Figure 1.

Table 2

Unpacked NOS-CI and embedded assessment

教學活動 NOS-CI 關鍵概念 嵌入式評量問題

2-1 水能溶解多少鹽 (2-3 節)

3-2-0-2 察覺只要實 驗的情況相同，產生 的結果會很相近

誤差 共識 可複製性 溶解量

6. 根據你蒐集的數據，比較各組的實驗 結果，你發現了什麼？

□每一組的實驗結果都完全相同

□每一組的實驗結果都相差很大

□每一組的實驗結果雖然不完全相同，

但卻很接近。

7. 假設你再做一次一樣的實驗，結果會 不會相差很大呢？

□會 □不會

8. 科學實驗的結果，都可以重複嗎？

□可以 □不可以

Interviews

Our participant teachers filled both VOSE-3 and VNOS in the professional development activities (詳細個案教師之資料請見總計畫及子計畫一). We then had follow-up interviews with individual participants. Interview questions and coding categories were developed to investigate participants’ views on NOS and NOS-CI and attitudes toward teaching NOS.

Classroom observations

Classroom observations focus on inquiry and explicit NOS instruction based on two reasons.

Firstly, the epistemological aspect of NOS is highly related to scientific inquiry. Secondly,

Akerson and Hanuscin (2007) have specified the effectiveness of teaching NOS through inquiry.

Therefore, any teaching that reflects inquiry and NOS is desirable for the participating teachers.

Selected items from Reformed Teaching Observation Protocol (RTOP) (Sawada et al., 2000) that are centred on inquiry and 10 additional items related to NOS are utilized to assess teachings.

Five ranking scales ranging from “Never Occurred” to “Very Descriptive” are used for the

protocol. The additional items are derived from classroom observations presented in the literature such as Akerson and Volrich (2006). Particular attention is paid to teachers’ vocabularies,

questions, and classroom discussion. Among the 10 NOS items, six belong to propositional knowledge:

1. The lesson incorporated NOS-CIs into explicit lesson objectives.

2. The lesson involved fundamental concepts of the NOS-CIs.

3. Please circle the NOS concepts discussed in the lesson: testable, predictable, repeatable, durability of scientific knowledge, tentativeness of scientific knowledge, consistency, limitation of science, objectivity, accuracy, factors that influence observations, what science is, what scientists do, and what scientists think.

4. The lesson highlighted the interrelationships between fundamental concepts of the NOS-CIs.

5. The teacher had a solid grasp of the content of the NOS-CIs.

6. Connections between the content of the NOS-CIs and real world were explored and valued.

Four items are related to the procedural knowledge:

1. The teacher made explicitly the connection between observations/empirical evidence and arguments/propositions.

2. The teacher brought students on evaluating the trustworthiness of propositions/arguments.

3. The teacher brought students to reflect on the similarities and differences between class activities (inquiry or discussion) and scientists’ investigation.

4. Students learned the content of the NOS-CIs from their inquiry.

結果與討論（含結論與建議）

Reliability and Validity of SVNOS

Test 1. The survey and interview data demonstrated that the students had a good

understanding of the items. For most items, less than 4% of the subjects either did not response or responded that they did not understand. Only one of the original 83 items had higher than ten percent of students who expressed a problem of understanding. The scale passed the

Kaiser-Meyer-Olkin (KMO) (value=.87) and Bartlett’s tests (p<.001) and were appropriate for EFA. Items not belonging to any specific factor were deleted. Some dimensions such as scientific knowledge is published, testable and culturally and socio-embedded were not found as originally designed. The constructs resulting from the EFA and sample items for each construct are presented in Table 1. The Cronbach’s alphas ranged from .67 to .84 (Table 3). The overall alpha was .85.

The first factor, theory-ladenness, included the influence of scientists’ thinking, experiences, knowledge, desires and resources on their observations. However, all reverse items of

theory-ladenness such as “If they [scientists] make different observations, someone must be

wrong” fell into the coherence/objectivity factor. These items stressed the consistency of observations made by different scientists and starting from scratch. They were in accordance with the objective image of science. Furthermore, two items that were originally designated for dimensions of scientific methods and diversity of thinking shifted to this construct. One was that there are many possible methods to solve a science problem. The other one was that different people may have different ideas when doing an experiment.

The second factor referred to the use of creativity and imagination in scientific investigation.

Scientists draw on their creativity and imagination to design experiments and interpret data.

Like artists, they need creativity and imagination for invention, innovation and inspiration.

Thirdly, the tentativeness factor incorporated a variety of reasons that students offered with regard to changes to scientific knowledge, i.e., scientific knowledge could be changed by innovative methods, technology, reinterpretation of data and new worldviews. Similar to the theory-ladenness factor, the factor analysis ruled out all negative items. Three of the negative items, which highlighted the well-proved, useful and cumulative features of scientific knowledge, shifted to the durability category.

Durability was the fourth construct. Initially this dimension focused on how scientific knowledge is established. For example, it is recognized and accepted by the science community and supported by measurements that are as precise as possible. Thus, it is not easily changed.

Afterwards, this construct incorporated three negative items from the tentativeness dimension, which stressed old theories being irreplaceable. The fifth factor was related to the consistency of experimental results, coherence of explanations and the objectivity of science. Scientific knowledge is framed as universal, not bound to society and culture. Experimental results should be consistent if the same materials and methods are used. Scientific explanations are based on evidence. Moreover, when scientists explain the results of an experiment, they have to make other scientists agree with them. This factor also involved some misconceptions such as there is one common scientific method, or observations should start from scratch.

The last two factors were originally constructed as one dimension to portray a

non-gender-biased contribution and engagement in science. However, the factor analysis of the empirical data entailed a split of the dimension into two factors, each about girls’ and boys’

engagement in science. The 7 first-order factors could explain both the upper grades and middle

grades’ data sets well, as shown by SEM (

.

Table 3

Number of items and reliability of the constructs of SVNOS NOS aspect No. of α

items Test 1 Test 2

Sample item

Theory-ladenness 9 .72 .70 Everyone has his/her own ideas and wants to prove them when making observations.

Scientists’ observations are influenced by their life experience and knowledge.

Creativity and imagination

6 .75 .73 Scientists need creativity and imagination to design and do experiments.

I believe that scientists work like artists. They need creativity and imagination.

Tentativeness 9 .77 .81 The advance of technology may lead to new findings and then it will change.

Scientific knowledge changes because people continue to change their views about the world and come up with new ideas.

Durability 6 .67 .77 Scientific knowledge which has been accepted by most people will not be replaced easily.

Scientific knowledge will not be replaced, because scientists have used it to get to the moon.

Coherence and objectivity

11 .79 .74 Experiments using the same materials and procedures will have exactly the same results.

There is only one method and one set of steps to do an experiment.

Science for girls 3 .84 .86 Girls have talent for scientific research.

Science for boys 3 .83 .87 Boys have talent for scientific research.

Total 47 .85 .85

Test 2. In the second test, eight of the 47 items had between 4 and 8 percent of data missing,

including no response and not understanding. The other items had less than 4% of the data

missing. The Test 2 data also passed the KMO (value=.87) and Bartlett’s tests (p<.001) for EFA.

Tests 1 and 2 support the same factor structure and measurement model. The 7 first-order factors could explain both the fifth and sixth grade data sets well, as shown by SEM (

.

The Cronbach’s alphas of the subscales were higher than those in Test 1 and ranged between .70 and .87 (Table 3). These values are satisfactory when compared with the existing NOS instruments. For example, the Beliefs about the Nature of Mathematics and Science subscale for preservice teachers resulted in .76 (McGinnis, Watanabe, Shama, & Graeber, 1997).

Meichtry (1992) administered the Modified Nature of Scientific Knowledge Scale (MNSKS) to

6

, 7

, and 8

graders and achieved alphas between .45 and .60 for the subscales and .77 overall.

As cited in Meichtry (1993), for secondary school and college student populations, Test on Understanding Science (TOUS) had .52 to .58 for its subscales and .76 in total; Nature of Science Scale’s (NOSS) split-half reliability was .72; and the alpha values of Nature of Scientific

Knowledge Scale (NSKS) were between .65 and .89, depending on the grade level. When NSKS was administered to 12

graders and college students, its alpha values were above .80.

However, for grades 10/11 and 9, it dropped to .74 and .65 respectively. Consequently, we concluded that the reliability of SVNOS for elementary school students was acceptable.

Summary. This study followed a development process for items that ensured an inherent

validity for the instrument and the trustworthiness and authenticity of the data, i.e., reliability (Aikenhead & Ryan, 1992; Chen, 2006; Erlandson, Harris, Skipper, & Allen, 1993; Rubba, Schoneweg Bradford, & Harkness, 1996). Furthermore, the consistency across constructs and measurements is high in SVNOS. The empirically-based approach yields a highly readable, valid and reliable questionnaire.

Model of Elementary Students’ NOS Views

The first-order factors were interrelated to some degree (Table 4). The correlation matrix showed that they could be clustered into three groups. Tentativeness (F1),

creativity/imagination (F2) and theory-ladenness (F3) had correlation coefficients ranging

from .49 to .59, p<.001. Durability (F4) and coherence (F5) as well as science for girls (F6) and science for boys (F7) were moderately correlated, r=.48 and .47, p<.001. Moreover, higher order factor analysis suggested three second-order constructs, G1-G3, corresponding to these three clusters of factors. F1-3 constructed a second-order factor about diversity of thinking and scientific methods, G1. Similarly, the F4 and F5 constructs could be interpreted by a

second-order factor about the objective aspects of science, G2. Furthermore, F6 and F7 had a higher-order factor, namely gender-free, G3.

Table 4

Correlations between factors and grade level

F1 F2 F3 F4 F5 F6 F7 grade

F1: theory-ladenness -

.16

-.09

.16

.20

-.02 F2: creativity -

.15

-.08

.16

.23

.02

F3: tentativeness - .13

.00 .17

.22

.05

F4: durability -

.14

.16

-.13

F5: coherence - .15

.13

-.12

F6: science for girls -

F7: science for boys - -.10

Note. ^**p<.001, ^*

p<.01

Absolute values higher than 0.40 are shown in boldface.

Explanatory power and factor loadings were used to determine whether the first- (Fs) or higher-order (Gs) factors should be retained. The Schmid-Leiman solution (SLS) provided additional insights into the factor structure by splitting up the direct loadings of variables on Gs and Fs. To determine the nonoverlapping contributions of the different levels of factors, the sums of squared loadings of corresponding items were calculated from SLS (Gorsuch, 1983;

Wolff & Preising, 2005). Furthermore, in order to accurately interpret the relationship between the observed variables and factors, the percentage of variance explained was derived from

standardizing these indexes by the total sum of squared loadings for each second-order factor (G) and the associated first-order factors (Fs). For example, the last six items in Table 2 about gender stereotypes belonged to F6, F7 and G3. The loadings of these items on F6, F7 and G3 were squared and then added by factor, resulting in 1.49, 0 and 2.44 for F6, F7 and G3

respectively, and totally 3.93. Standardized by the total sum of squared loading 3.93, F6 and G3 explained 37.88% and 62.12% of the variance. The general factor G3 explained all the variance of F7—science for boys. However, specific elements of F6—science for girls, accounted for 37.88%, a notable amount of variance.

In addition to the explanatory power, the factor loadings also demonstrated that the three items associated with girls’ engagement in science were more closely related to F6 than to G3.

The other three items about science for boys directly measured G3. F6 and G3 were evidenced hitherto. However, it could be that G3 replicated F7—science for boys. This is conceivable in a factor structure that involves correlated lower order factors and a higher order factor. SLS computes individual contributions of factors of different levels. It does not assume a correlation between factors of different levels; rather, it looks into the direct relations between items and higher order factors. The direct relations improve the interpretation of higher order factors, which otherwise would rely on the interpretation of lower order factors (Wolff & Preising, 2005).

In the present case, the general factor G3, ideally standing for gender-free, mainly rested on the variables related to the male stereotype in science. Therefore, a more reasonable model is that two correlated first-order factors, F6 and F7, exist, instead of a hierarchical relationship between F6 and G3. In other words, students’ beliefs about gender stereotype girls’ and boys’

engagement in science are correlated. Likewise, objectivity of science (G2) replicates durability (F4) and is redundant.

For F1, F2, F3 and G1, SLS showed that all of the corresponding items had higher loadings on G1 than on the first-order factors, and appeared to reflect G1. Moreover, F1, F2 and F3 accounted for relatively low percentages of variance explained in observed variables

(8.10-19.11%) compared with G1 (64.42%). Both the factor loadings and explanatory power favored the general measure G1—diversity over the accurate measures F1, F2 and F3.

Nevertheless, as there is no definite critical value of explanatory power for a factor being

considered as meaningful, the selection of level of factors also depends on the purpose of the

study. For many science teachers and researchers, first-order factors and their interaction may

provide more details for curriculum and instruction decisions. Additionally, it is possible that

when students have formed more concepts about the theory-ladenness of observations and data

interpretations, the use of creativity and imagination in science and the tentativeness of scientific

knowledge, the first-order factors may play a more definite role in the factor structure.

Therefore, we suggest retaining the first-order factors.

In sum, the data suggested a model of seven factors: theory-ladenness; creativity and

imagination; tentativeness; durability; coherence and objectivity; science for girls; and science for boys. The first three factors could be explained by a higher order factor, diversity.

Elementary Students’ Development of NOS Views

The preliminary results of Test 1 provided some information about how students develop their NOS views from the third to sixth grades. The data of the 47 items elicited from the item and factor analyses were taken into account. The average scores by grade, and their

comparisons based on Test 1 are displayed in Table 5. the students retained the same view about diversity (or tentativeness, creativity and theory-ladenness) throughout the four grades.

However, their conceptions about durability, coherence, and science for girls and boys were slightly and negatively correlated with their grade level, r=-.13, -.12, -.08, -.10, p<.01, respectively.

Table 5

Comparison of students’ NOS views NOS aspects Grade 3

(n=261)

Grade 4 (n=303)

Grade 5 (n=441)

Grade 6

(n=134)

Theory-laden 4.12 4.06 4.07 4.05 .69 .558 .002 NA

Creativity 3.95 3.83 3.91 3.95 1.98 .115 .005 NA

Tentativeness 3.77 3.75 3.82 3.85 1.35 .254 .004 NA

Durability 3.31 3.21 3.05 3.04 8.13 .001 .021 Grade 3>4,5,6 Coherence 2.72 2.47 2.49 2.42 8.40 .001 .022 Grade 3>4,5,6 Sci. for girls 3.52 3.41 3.36 3.25 2.36 .070 .006 Grade 3>6 Sci. for boys 3.76 3.65 3.45 3.56 5.60 .001 .015 Grade 3>5

neither agree nor disagree (=3), somewhat agree (=4) and strongly agree (=5).

Theory-ladenness. All grades of students demonstrated a consistent view of theory-laden

observations. The average score of this aspect was the highest among all the topics. In Taiwan,

science is integrated with social studies and arts in the first and second grades. Students start to

receive formal science instruction from the third grade. The third grade curriculum is mainly

about making observations. The grade 3-6 science curriculum incorporates many hands-on

activities. Although the students did not know the term theory-laden, the questionnaire data

revealed that they had experience and knowledge of how observations are made, including

thinking about and setting up specific objectives before observations, following a predetermined

table for recording observations, looking for differences between pre- and post-observations,

attempting to confirm what is in mind, and the influence of society, culture, training and personal

experiences. Curriculum documents and previous studies usually do not involve the issue of

theory-laden observation at the elementary level (e.g., AAAS, 2007). Nevertheless, the results

revealed that these elementary students were ready to discuss the issue based on their experiences.

Creativity and imagination. Like the theory-laden aspect, students had a fairly stable view

of the use of creativity and imagination in science. The theme is commonly found in the few studies that have involved elementary and middle school students, whereas most studies focus on pre-/in-service teachers and high school students’ NOS views. The findings of first to seventh graders’ conceptions of NOS prior to any teaching intervention of NOS are summarized in Table 5. The results vary with the populations and research methods. Huang, Tsai and Chang (2005) surveyed over 6,000 5

and 6

graders in Taiwan using a combination of items related to

creativity, imagination and tentativeness. They reported that most students recognize the invented and changing nature of science. Similarly, 60-70% of our participants were positive about the items of this subscale. They believed that scientists use imagination and creativity to design experiments, to explain abstract concepts and to get inspiration from nature. A higher percentage of students in Taiwan recognized the role of creativity and imagination in science compared with the findings of research from the US.

Akerson and Abd-El-Khalick (2005) pointed out that everyday language has imposed a dreamy, unreal sense on the definitions of creativity and imagination. As a result, students do not tend to make connection between science and creativity/imagination. Students in Taiwan might perceive creativity and imagination differently. Particularly, they often related creativity to innovation in their written responses and interviews. In fact, the Ministry of Education has promoted creativity education since the last decade. The White Paper on Creative Education has a vision of making the Republic of China (the official name of Taiwan) a Republic of

Creativity (Ministry of Education, 2002). Funds have been allocated to foster creative learners, creative teachers, creative curricula and creative schools. An effort has been made to intertwine the element with every aspect of schooling. Consequently, the participants of this study

appeared to be more positive about the role of creativity and imagination in scientific investigation.

Tentativeness. Over half of the subjects agreed that scientific knowledge may change due

to new methods, technology and worldview (Table 5). Less than one fifth thought that scientific knowledge does not change. Three of the previous studies, Kang et al. (2005), Akerson et al.

(2005) and Huang et al. (2005), had similar results. However, Akerson and Volrich (2006) and Khishfe (2008) reported contradictory findings. The disparity could be caused by curriculum, culture and research methodology. Further research is needed to understand the cause.

Table 5

Summary of students’ pre-instruction views of the use of creativity and imagination in science and the tentativeness of scientific knowledge

Creativity/imagination Tentativeness Study Participant

Informed view

Naïve view

Science changes

Science

does not

change

Akerson et al. (2006) 14 1

graders 43% 57% 36% 64%

Akerson et al. (2005) 23 4

graders 9% 91% 70% 17%

Khishfe (2008) 18 7

graders 22% 50% 17% 72%

Kang et al. (2005) 534 Korean 6

graders NA NA 69% 16%

Huang et al. (2005) 6167 Taiwanese 5

-6

graders

An average of 4.36 on a 5-point Likert scale that combines the two factors.

Current study 1139 Taiwanese 3

-6

graders

60-70% 7-16% 50-74% 6-17%

Durability and coherence. Unlike the diversity aspect, students had lower scores on the

durability and coherence aspects. These factors involved some items that address

misconceptions. An increasing number of students expressed disagreement on those items as they moved toward the 6

grade. For example, on the item “Scientific knowledge will not be replaced because it has been proven by experiments and explanations”, the percentages of third to sixth graders who took the disagreement side were 34, 47, 49 and 53%, respectively. On the contrary, decreasing percentages (28, 23, 21 and 16%) agreed with the statement. Likewise, fewer sixth graders (39%) believed that scientific experiments follow fixed steps and method, compared with the third graders (55%). Students significantly modified their misconceptions.

Gender stereotypes. Students tended to deem that boys fit science better than girls, which

is consistent with the common stereotype. Their views on science for girls/boys were negatively correlated with grade levels. The more science they had, the less they felt that science is for everyone.

The results suggested a seven factor model to assess and explain young learners’

conceptions of NOS. Some items did not follow the dimensions proposed by the literature or researchers. In other words, students held a different structure of thought than the experts. An understanding of their perspectives can help provide recommendations for development of curricula and instruction. Students may develop their epistemological views along two lines.

One is coherence and objectivity, and the other is tentativeness and diversity of thinking.

Students view scientific knowledge as being coherent, empirically-based, objective, fixed steps of scientific research and so durable that it almost never changes. On the other hand, they perceive science as tentative, creative, imaginative, theory-laden and requiring diverse thinking. Students construct NOS views along these two lines. It is difficult for them to form a balanced view that is internally consistent.

The preliminary results indicated that, with no explicit NOS instruction, the students did not significantly alter their informed views, but slowly amended their misconceptions from the third to sixth grades. This tendency may change once the students get to secondary school as the curriculum, instruction and classroom settings are dramatically different from those in elementary school. Future research may look into the development of these two lines of thinking in older children, which would provide valuable information for curriculum design.

Several recommendations can be drawn from this study. First of all, science educators and

teachers should pay attention to the presentation of how the abovementioned two lines interweave

to produce science, so that students may construct a comprehensive view of NOS. Secondly, elementary students have experiences and the background to discuss the theory-laden nature of observation, which has previously been neglected at the elementary level. Thirdly, creativity and imagination are important elements in science. In school, they should not be bound to the arts and literature alone, but should be promoted in science learning. Students may therefore modify their definitions of creativity and imagination, and realize their role in science. Finally, scientific knowledge is consistent, coherent and tentative. The consistent aspect is highlighted in reform documents for young learners. However, it is challenging for teachers and curriculum designers to present this consistency and avoid building or encouraging concepts of a universal scientific method and unchangeable scientific knowledge. A more effective strategy of

instruction would be to generate discussions of NOS from multiple facets, rather than focusing on separate NOS tenets. Professional development activities may help teachers to gain a balanced view of NOS so that they are able to prompt a multi-faceted discussion with students.

This study has several limitations. We centered on epistemological views and common stereotypes, leaving ontological and sociological aspects unexplored. Many important factors in the enterprise of science such as class and ethnicity were not included in this study. The gender stereotype deserves further discussion. Moreover, this instrument can be used for fourth graders and older students. In this study, the fourth graders had a 5-10% response of not understanding on less than one tenth of the items. However, when it was administered to third graders at the end of the second semester, nearly half of the items received a relatively high percentage (5-15%) response of “do not understand”. Researchers would have to triangulate the results if it is to be used for students in lower grade levels or with limited reading comprehension. Finally, the structure model may be varied for different populations. The structure, such as the lower and higher order factors, may change for secondary school students. It will be interesting to investigate the development of the factor structure.

More results and discussions regarding the observation protocol, embedded assessment and interviews please may be found in the report of the integrated project.

初步研究成果總結如下：

(1)發展並測試學生科學本質問卷

(2)修改 VOSE 教師科學本質問卷。作為每次教師研習之前後測。

(3)教師晤談題目。我們根據第一次訪談的結果精緻第三階段的教師研習，特別針對教師們 較不清楚的 NOS 議題深入介紹。

(4) 完成課程教材「溶解」單元之嵌入式評量。

(5)RTOP 和外加之 NOS 觀察題項。用以評析個案教師之教學，並提供回饋。

(6)研討會論文：與團隊共同發表有關教師之科學本質信念、嵌入教材之科學本質評量，以 及評量學生之科學本質觀的模型等論文。

(7)期刊論文：發表教師科學本質問卷，以及國小學生之科學本質觀的結構。

(8)國際交流：

„ 在 97 年 3 月 8 日之素養活動與科學學習研習，Brian Hand 教授蒞臨，全程指導。

„ 與 Brian Hand 教授以 Skype 聯繫。並在 NARST2008 會議檢討 97 年 3 月之教師

專業成長研習，以及 97 年 5 月研習之內容與重點。

„ 與美國、大陸、土耳其學者合作開發科學本質評量工具。

„ 在 NARST 會議與 DeBoer, Liu, Barnett,

等國際學者討論統計法和教師 專業發展。

參考文獻

Abd-El-Khalick, F., & BouJaoude, S. (1997). An exploratory study of the knowledge base for science teaching. Journal of Research in Science Teaching, 34(7), 673-699.

Abd-El-Khalick, F., & Lederman, N. G. (2000). The influence of history of science courses on students’ views of nature of science. Journal of Research in Science Teaching, 37(10), 1057-1095.

Aikenhead, G. S. (1988). An analysis of four ways of assessing student beliefs about STS topics.

Journal of Research in Science Teaching, 25(8), 607-629.

Aikenhead, G. S., & Ryan, A. (1992). The development of a new instrument: “Views on Science-Technology-Society” (VOSTS). Science Education, 76(5), 477-491.

Akerson, V. L., & Abd-El-Khalick, F. S. (2005). "How should I know what scientists do?—I am just a kid": Fourth-grade students' conceptions of nature of science. Journal of Elementary

Akerson, V. L. & Hanuscin, D. L. (2007). Teaching nature of science through inquiry: Results of a 3-year professional development program. Journal of Research in Science Teaching, 44, 653-680.

Akerson, V. L., & Volrich, M. L. (2006). Teaching nature of science explicitly in a first-grade internship setting. Journal of Research in Science Teaching, 43(4), 377-394.

American Association for the Advancement of Science. (1993). Benchmarks for science literacy:

Project 2061. New York: Oxford University Press.

American Association for the Advancement of Science. (2007). Atlas of science literacy, Vol 2.

New York: Oxford University Press.

Chen, S. (2006). Development of an instrument to assess views on nature of science and attitudes toward teaching science. Science Education, 90(5), 803-819.

Chiu, M.-F., & Kao, H.-L. (2004). A research of the integration of the history of science into science teaching to promote elementary school children’s viewpoints of the nature of science. Journal of National Taipei Teachers College, 17(1), 183-214.

Chiu, M.-F., & Kao, H.-L. (2006). An exploration of effect of integrated teaching about the history of science upon elementary school children’s viewpoints of the nature of science.

Chinese Journal of Science Education, 14(2), 163-187.

Clough, M. P., & Olson, J. K. (2004). The nature of science: Always part of the science story. The

Dogan, N., & Abd-El-Khalick, F. (2008). Turkish grade 10 students' and science teachers'

conceptions of nature of science: A national study. Journal of Research in Science Teaching,

45(10), 1083-1112.

Clough, M. P. (2006). Learners’ responses to the demands of conceptual change: Considerations for effective nature of science instruction. Science & Education, 15(5), 463-494.

Duschl, R. A., & Gitomer, D. H. (1997). Strategies and challenges to changing the focus of assessment and instruction in science classrooms. Educational Assessment, 4, 37-73.

Elder, A. D. (2002). Characterizing fifth-grade students’ epistemological beliefs in science. In B.

K. Hofer & P. R. Pintrich (Eds.), Personal epistemology: The psychology of beliefs about

Erlandson, D. A., Harris, E. L., Skipper, B. L., & Allen, S. D. (1993). Doing naturalistic inquiry:

A guide to methods. Newbury Park, CA: Sage.

Gerretson, H., & Golson, E. (2005). Synopsis of the use of course-embedded assessment in a medium sized public university’s general education program. The Journal of General

Gignac, G. E. (2007). Multi-factor modeling in individual differences research: Some

recommendations and suggestions.

Good, R., Lederman, N., Gess-Newsome, J., McComas, W., & Cummins, C. (2000). Nature of

paper presented at the annual international meeting of the Association for the Education of Teachers in Science, Akron, OH.

Gorsuch, R. L. (1983). Factor analysis (2

ed.). Hillsdale, NJ: Erlbaum.

Green, C. W., Middleton, S. G., & Reid, D. H. (2000). Embedded evaluation of preferences sampled from person-centered plans for people with profound multiple disabilities. Journal

Hayton, J.C., Allen, D.G., & Scarpello, V. (2004). Factor retention decisions in exploratory factor analysis: A tutorial on parallel analysis. Organizational Research Methods, 7(2), 191-205.

Huang, C.-M., Tsai, C.-C., & Chang, C.-Y. (2005). An Investigation of Taiwanese Early Adolescents' Views about the Nature of Science. Adolescence, 40(159), 645-54.

Kang, S., Scharmann, L. C., & Noh, T. (2005). Examining students’ views on the nature of science: Results from Korean 6

, 8

, and 10

graders. Science Education, 89(2), 314-334.

Khishfe, R. (2008). The development of seventh graders' views of nature of science. Journal of

Kim, J.-O., & Curry, J. (1997). The treatment of missing data in multivariate analysis.

Sociological Methods & Research, 6(2), 215-240.

Kitcher, P. (1998). A plea for science studies. In N. Koertge (Ed.), A house built on sand:

Exposing postmodernist myths about science (pp. 32-56). New York: Oxford University

Press.

Kourany, J. A. (1998). Scientific knowledge: Basic issues in the philosophy of science. Belmont, CA: Wadsworth Publishing Company.

Leach, J., Hind, A., & Ryder, J. (2003). Designing and evaluating short teaching interventions

about the epistemology of science in high school classrooms. Science Education, 87(6),

831-848.