Modified guided-discovery methods in physics laboratories: Pre-service teachers’ conceptual and procedural knowledge, views of nature of science, and motivation

Abstract Although laboratory work is considered a significant factor in helping students develop their science process skills and cultivate alternative knowledge construction, there are still gaps in the selection, integration, and implementation of the basic elements in physics laboratories. In this study, three modified guided discovery and conventional methods in integration with different forms of laboratories, the explicit approach of process skills, and the implicit approach to the nature of science were implemented. This study aimed to explore pre-service teachers’ views towards nature of science (NOS), pedagogical and forms of laboratory orientation, practice of science process skills (PS), and motivation. The groups were also compared in terms of the dependent and selected covariate variables. This study employed a tandem design Phase III with a quasi-experimental approach. Purposive and random sampling were used to select study participants. It was found that pre-service physics teachers had naïve views towards NOS and PS before and after the intervention. However, after the intervention, there were changes in pedagogy and forms of laboratory orientation, PS practice, and motivation. The groups were also significantly different in conceptual and procedural knowledge and motivation, but similar in views of NOS. In addition, the groups were not significant for covariate variables. Generally, this study indicates that modified guided-discovery methods in physics laboratories have made differences in procedural and conceptual knowledge, motivation, and practicing process skills. It was suggested that physics teachers and physics laboratory curricula employ a variety of learning models and integrate basic components to assist and improve students’ learning.


Introduction
Although science educational policies and curricular materials state excellent learning outcomes for student learning, science education studies have shown that many students achieve fewer outcomes.This implies that students are less sufficiently learning the basic concepts, procedures, and nature of science (Baloyi et al., 2017).This is also a concern for many teachers (Blanchard et al., 2008;Ramarian, 2016).According to Blanchard et al. (2008), neither students nor schoolteachers have clear ideas about how science operates and how scientific knowledge is constructed.Teachers may have gaps because they are limited and affected by the teaching materials they use in schools or colleges (Adisu et al., 2021a).
To minimize the problems in science/physics education in terms of students' learning outcomes and motivation, research findings recommend curriculum materials (textbooks, modules, and laboratory manuals) in science education to integrate and implement basic components of science.The main basic components are science process skills, concepts, nature of science, alternative pedagogy, forms of laboratory, and assessment mechanisms (Hofstein & Lunetta, 2003;McDermott, 2013).In science education, appropriate integration and implementation of basic components open the opportunity for students to gain knowledge about the real picture of what scientists do in investigating scientific findings, such as theories, laws, and principles.In addition, such integration helps students understand the nature of science and process skills, and it also motivates students towards science learning (Hodson, 2002).Despite their significance, the types of basic components and models for their selection, integration, and application in science education are less widely acknowledged internationally.That is, there is a lack of focus on the types of basic components and forms of selection, integration, and implementation to construct alternative knowledge in science education (Adisu et al., 2021b).Thus, this area needs an alternative model of learning that guides the selection, integration, and implementation of basic components in science education.
In science education, laboratory work is believed to play a prominent role in practicing science process skills, cultivating alternative knowledge construction, understanding science, and exposing students to different forms of learning environments (Sudarmani et al., 2018;Zudonu & Njoku, 2018).In addition, it is a student-centered method that requires learners to be active (Shimeles, 2010).In laboratory work, students manipulate real objects and develop the arts of experimentation, analytical skills, and conceptual learning.In addition, it helps students understand the basis of knowledge in physics (science), develop informed views on the nature of science, and motivate (Baloyi et al., 2017;National Research Council, 2006).Due to this, with the absence of laboratory work in science education, the quality of students' science learning is not expected (Sudarmani et al., 2018).Thus, school and college curricula are expected to integrate and implement laboratory work appropriately to support students' science learning (Adisu & Abebaw, 2021;Daniel et al., 2023).
There are many trends in science education laboratory work.However, most of them have focused on the implementation of selected pedagogies and their impacts on students' learning outcomes and motivation (Baloyi et al., 2017;Leung et al., 2017).Others have focused on comparing the impact of different forms of laboratories on students' learning outcomes and motivation (Banchi & Bell, 2008;Bell, 2008;Holmes et al., 2017;Parreira & Yao, 2018;Ramarian, 2016, Winning, 2005;Winning, 2011).The implicit or explicit approach of nature of science (NOS) and process skills (PS) in science laboratory work is mostly used area of studies in science laboratories (Baloyi et al., 2017;Kalman et al., 2018).A few studies have been conducted on the critique of laboratory work and content analysis of science/physics laboratory materials (Blosser, 1980;Hofstein & Lunetta, 1982, 2003;Shimeles, 2010;Singh, 2014).In all the above studies, there were gaps in the selection, integration, and implementation of the aforementioned basic components in science/physics laboratory work.Hofstein andLunetta (1982, 2003) and Singh (2014) conducted content analyses on laboratory work, which revealed that the majority of studies have failed to adequately explain students' abilities and attitudes, as well as to incorporate basic components in laboratory experiments.They further argued that even the development of standardized achievement tools has been ignored to measure students' laboratory outcomes.These tools primarily emphasize testing factual or conceptual knowledge; they do not investigate teachers' methods of instruction concerning laboratory tasks.Furthermore, content analysis conducted by Shimeles (2010) in physics laboratory materials (manuals) indicated that there is ignorance of affective variables, such as attitudes and interest, and discrepancies between learning goals and actual learning in laboratory classes.In addition, the materials were more content-centered and designed for conventional methods and confirmatory laboratory tests.In addition, as noted by Blosser (1980), most laboratory studies have no conceptual framework that guides them.
In addition to the above-mentioned limitations in science laboratory work, the overviews of laboratory curricula used in colleges of teachers' education in Ethiopia indicated that there are limitations in the material development (laboratory manuals) process and its implementation in the college of teachers' education (Adisu et al., 2021a;MOE, 2018).The main identified gaps in physics laboratory manuals of the College of Teacher Education indicated that there is less unclear or nonexplicit integration of contents, process skills, the nature of science, and allocation of appropriate pedagogy, forms of laboratory in laboratory manuals, and assessment used to measure laboratory sessions.It was found that there is a lack of a clear model of learning that guides the selection, integration, and implementation of pedagogy, forms of laboratory, and assessment mechanisms in physics laboratory curricula, as well as in previous studies (Adisu et al., 2021a;Daniel et al., 2023).Furthermore, there is a paucity of criteria used to select and derive dependent and covariate variables in previous studies related to science laboratory tasks (Adisu et al., 2021b).Such gaps lead to a mismatch between the process of instruction and the objectives to be attained (Badri & Shri, 2013;Hofstein & Lunetta, 2003;Shimeles, 2010).It has been reported that students (pre-service teachers in the college of teachers' education) do not sufficiently learn the basic concepts, procedures, and nature of science in a science laboratory (Baloyi et al., 2017).Hence, to minimize these gaps, this study was conducted at the College of Teachers' Education Physics Laboratory.
It was found that there is a need to investigate the impact of different learning models on students' views and learning outcomes in terms of different integration of basic components to help them be equipped with knowledge and skills in constructing scientific knowledge.The results from Adisu et al. (2021a) and Adisu et al. (2021b) revealed that scientific knowledge can be constructed in science/physics laboratories by integrating some of the basic components, such as the structured form of content/curricula and controlled form of laboratory, structured curricula/ content and uncontrolled form of laboratory, semi-structured curricula and controlled form of laboratory, and open contents/curricula and open form of laboratory.Hence, the studies under the same project developed and implemented five models (structured guided-discover (SGD), semistructured guided discovery (SSGD), Scaffolding guided discovery (SCGD), free discovery, and traditional) of learning (see Table A1 in the appendix) that can be used in science laboratory work and can serve as alternative models to derive and select dependent variables and covariates in science laboratory work.Under the same project, the intervention impact of some models on some selected dependent variables within group comparison was reported by Adisu et al. (2021a), and the association of pedagogies and some selected covariates was reported by Adisu et al. (2021b).In this study, the intervention impacts of four selected alternative models on students' learning outcomes and motivation among groups compared with the control of covariates and exploration of students' views of NOS, PS, pedagogical, and forms of laboratory orientation were reported.The following research questions guided the study: (1) What are the pre-and post-conceptions of pre-service physics teachers' views about NOS and PS, pedagogical and forms of laboratory orientation, practice of process skills, and motivation?
(2) Is there a significant mean score difference among groups in terms of covariates of pedagogical and forms of laboratory orientation, PS practice, and overall academic performance after intervention?
(3) Is there a significant mean score difference among groups in terms of conceptual and procedural knowledge, NOS and PS, and motivation with control of covariates after the intervention of the study design?

Design of the study
In this study, the same study design was used as reported by Adisu et al. (2021b).The study was a tandem design phase (III) using a quasi-experimental approach (Campbell & Stanley, 2005).This design was selected to compare the three paired matched treatment groups and one control group.Self-determination (Deci, 1975;Deci & Ryan, 1987) and social constructivism (Vygotsky, 1978) theories were used to guide the study.Self-determination theory is used to modify the guided discovery method, whereas the social-constructivism theory of learning is used to support classroom intervention.In the self-determination and teaching-learning process, students need proficiency, independence, and connection with the teacher.This theory supports the current study (modified guided discovery methods) in such a way that it informs the development and implementation of the three modified guided discovery methods used in this study.That is, it requires students' proficiency, independence, and connection, which commands the attention of the instructor.Therefore, the theory has direct implications for guided discovery methods.On the other hand, social constructivism served as a guide for classroom intervention in a balanced student-teacher approach, and knowledge is considered mainly to be constructed rather than acquired in the three modified guided discovery methods.The theory has a contribution in giving a clue as knowledge construction via social interaction in science laboratory work that supports five models of learning (Adisu et al., 2021b)).According to the theory of social constructivism, learning in science classes is due to social discourse; hence, students acquire and/or build alternative knowledge (Vygotsky, 1978(Vygotsky, , 1986)).

Method of the study
In this study, a mixed method was employed.Qualitatively, the study used transcendental (empirical) phenomenology and quantitative comparative studies were used (Creswell & Clark, 2007;Creswell & Poth, 2013).Transcendental (empirical) phenomenology was used to explore and describe various conceptions of the phenomenon by the subject of studies (Donalek, 2004;Mariano, 1990).In this study, empirical phenomenology was used to explore and describe students' views on NOS, pedagogical and forms of laboratory, practicing process skills, and motivation for the data collected using semi-structured tools regarding the variables when different pedagogy and forms of laboratory were implemented.Qualitatively explicated variables were students' views on NOS and PS, forms of laboratory and pedagogical orientation, and motivation.The procedures suggested by Hycner (1999) were implemented to explicate the data.The comparative study (MANCOVA) test was used to measure and compare the impact of the study design among groups in terms of dependent variables and covariates.It was used to compare students' views of NOS, practicing PS, motivation, and conceptual and procedural knowledge to what extent the implemented pedagogy made a difference among groups.

Population and samplings
In this study, the same population and sampling method used by Adisu et al. (2021a) were used.
To conduct this study, two colleges were selected: Arbaminch and Hossana College of Education.
The reasons for selecting the two colleges were that they were easily accessible by the researcher, availability of equivalent materials to conduct experiments selected, availability of experienced teachers in the college, and the availability of the two groups in each college.By purposive sampling, third-year physics students were selected among the groups taking the course because they had more experience in laboratory work than first-and second-year students.Random sampling was used to assign students (two groups in each college) to the control and treatment groups.The total number of students at the two colleges was 135.Of these, 112 students participated in both pre-and post-administration of the study variables.The comparison group (M = 21, F = 3), SGD (M = 23, F = 9), SSGD (M = 20, F = 7), and SCGD (M = 26, F = 3) participated.To qualitatively identify students' perceptions of NOS and PS, practicing process skills, pedagogical and forms of laboratory orientation, and motivation random sampling were used.From each group, 41% of the students were selected using random sampling for qualitative data.

Data gathering procedures and treatments
In this study, the same data gathering and treatments were conducted as those reported by Adisu et al. (2021a) and Adisu et al. (2021b).Before the intervention, the facility condition was checked and consent agreement was obtained.To conduct ethical research, the rules and regulations of Addis Ababa University were followed.A letter of authorization and cooperation was obtained from the university.Colleges were permitted to conduct research in natural classrooms.Written informed consent was obtained from all participants (college instructors).The recommended items included in a consent agreement participate in research and generally define the purpose of the research (without saying the key research question).In addition, the research procedures, the risks and benefits of the research, the voluntary nature of research participation, and the subject's (informant's) right to stop participation at any time were pre-informed to protect confidentiality.
Training was provided to the instructors who implemented the three modified guided discovery methods.The training focused on different models of learning, pedagogies used in physics laboratories, forms of laboratories suitable for pedagogy, presenting content and NOS, presenting questions and replying to the questions, and how to assess (both formative and/or summative).In addition, lesson plans and tools were developed and validated before the intervention, at the same level as students in other colleges.Before the intervention, a pre test was conducted.During the intervention phase, continuous formative and supportive feedback regarding the distribution of the checklist and reporting formats for each experimental work was provided to both students and teachers.To minimize within-group contamination before and after conducting the experiments, first-level data analysis and answers to selected questions were conducted and signed/marked by the class teacher.The title of the experiments to be done the next time was not informed to the experimental groups; rather, they were informed after they came to the class.Totally seven experiments were conducted within 10 weeks out of 16 weeks of the academic semester.The experiments conducted were Charges and methods of charging bodies, series and parallel DC Circuits, verification of Kirchhoff's law, determination of electrical resistance and Ohms law, internal resistance and EMF of dry cell, and investigation of induced EMF in solenoid.After the study design was implemented, a post test was conducted.In addition, compensation sessions were conducted for the treatment groups.The overall research procedure of this study is shown in Figure 1.

Data collection tools
After identifying the dependent variables and covariates according to the models reported by Adisu et al. (2021a) and Adisu et al. (2021b), tools were adapted to measure the study variables.To explore students' views on NOS and PS, semi-structured tools were adapted from Liang et al. (2006).The Tools used to measure conceptual and procedural knowledge were developed in the context of laboratory work according to Hofstein andLunetta (1982, 2003) and validated.Questions related to pedagogy and forms of laboratory orientation and practicing PS were developed based on the context of the study and validated.Furthermore, students' overall academic performance (CGPA) data were collected from the colleges' registrar's office.

Tool validation
The internal consistency (reliability) test of the conceptual and procedural tests was performed using the Kuder-Richardson KR20.A reliability test of the Likert-scale questions was conducted using Cronbach's alpha.The pilot test results were reported by Adisu et al. (2021a) and Adisu et al. (2021b), respectively.The reliability test indicated that all the instruments on the Likert scale were reliable and within the usable range.Cronbach's alpha of the test was within the acceptable range (α>0.7)(Nunnally & Bernstein, 1994).Cognitive test results were also within the acceptable range (Cortina, 1993).Validity tests of tools, such as the face, content, and construct validity of semi-structured and rated scale questions, were conducted by participating in two English and Amharic languages and three physics instructors in the colleges of teacher education.The criteria used were work experience (in college) and educational background (holders of a master's degree).In addition, to validate conceptual (10 items) and procedural (10 items), the phi coefficient (ϕ) was used by comparing three physics instructors in colleges with the researcher.The best correlation factor was 0.78.However, the items were modified based on the low correlation values.

Statistical analysis techniques
This study employed transcendental phenomenology (description of different views) and a descriptive comparative survey.According to Mariano (1990), transcendental phenomenology is used to explore the different conceptions of students.Based on this, students' views of NOS and PS, pedagogical and forms of laboratory, practice of PS, and motivation were explicated before and after the intervention.To identify the groups' conceptions, the procedures suggested by Hycner (1999) were implemented.The MANCOVA test was used to compare the groups after the intervention in terms of dependent variables with control of covariates.Before testing the hypotheses, a normality test (skewness test) was conducted.The result indicated that −1<skewness <1) (Hair et  al., 1998;Huck, 2012;Ramos et al., 2018) or, −1.96< skewness/std.error<1.96 and/or −2.58< skewness/std.error<2.58 (Ghasemi & Zahediasl, 2012).All data types of each group and the overall groups' data normality test indicated that the parametric test assumptions were fulfilled; thus, a parametric test was conducted.SPSS Version 22 was used to perform the analyses.

Variables of the study
In this study, the independent variables were the basic components, such as models of learning, forms of laboratory, pedagogies, content, NOS, and PS.The effect of two independent variables was considered as a covariate, and three independent variables were considered as dependent variables.Therefore, the combination of forms of the laboratory with contents, NOS, or PS provides a covariate called practicing process skills; a combination of forms of the laboratory with pedagogy to teach content, NOS, and PS yields another covariate called different level pedagogical orientation; and a combination of pedagogy with contents, NOS, or PS provides a covariate called mastering of contents, PS, and NOS.The combined effect of three independent variables-pedagogy, content, and forms of laboratory-provides overall learning outcomes and motivation (conceptual and procedural knowledge, understanding of NOS, and motivation).Hence, Conceptual and procedural knowledge, views of NOS, and motivation were selected as dependent variables, while practicing process skills, pedagogical orientation, and overall academic performance were selected as covariates.Table 1 depicts the manner in which the dependent variables and covariates were specified for the study.

Pre-conception about views of NOS and PS, pedagogical and forms of laboratory and motivation
In this section, research question one is answered.The data were analyzed using semi-structured questions to identify students' views on NOS and PS.In the explication of the semi-structured data, each group's view was categorized and described rather than interpreted.The objective of using this method was to explore students' views on NOS and PS, pedagogical and forms of laboratory orientation, and motivation, and to identify the nature of groups and misconceptions about the views of NOS and PS.The findings were triangulated using the quantitative findings in later sections.To identify themes, the procedures proposed by Hycner (1999) were followed.First, the students' responses to the questions were described without explanation or interpretation (bracketing).This was followed by demarcating the scope of concepts related to each question, code, and categorizing the same meaning/statements or concepts in one group, that is, clustering of units of meaning by collecting the same unit of ideas to form themes (Delineating units of meaning).Third, each semi-structured question under one theme was summarized, validated, and modified to avoid changing the meaning.Finally, general and unique ideas from all semi-structured questions were extracted to create a composite summary of thoughts and statements for each query.Themes were identified by the frequency count of similar ideas, concepts, words, and terms generated by the students.The results for the pre-conception students' views are presented according to the criteria used to measure the variables in the following section.The data preseted in appendix Table A2.

Observation and interpretation difference in science
Under this criterion, the findings indicate that in all groups, the majority of students (more than 60%) agreed that there was a difference between observation and interpretation among scientists for a certain phenomenon, except for the SSGD group.The reasons they thought for this response were that the difference between scientists in observation and inferences may be due to their background differences and their data gathering and observation techniques.This implies that initially, the majorities of the selected students in all groups were relatively similar and had wellinformed views on observation and interpretation in scientific research.However, approximately 40% of participants in each group had naïve views.

Changeability of theory and law in science
Under this criterion, the finding indicates that the majority of selected students in all groups thought that theory and law were changeable.They responded that if existing theories and laws are tested by other scientists in an advanced way, they may obtain stronger and better evidence, and errors could be obtained in the previous findings.As a result, theories and laws can be subjected to change.More than 54% of respondents in each group supported this idea.However, more than 45% of the participants had a naive view of each group.

Difference of theory and law
Under this criterion, the findings revealed that some students (about 30%) in each group perceived that laws are the steps/rules used to conduct scientific activities and are formulated by scientists.Some others (30%) also thought that the law was a finding of the research.Across all groups, the majority of the students had similar perspectives.However, in all groups, there were naive views on law in science.In addition, more than 15% of the students were uncertain about their answers.Concerning theory, more than 53% of the respondents from each group perceived that theory is tentative, testable, and a changeable idea of the scientist about observation.In addition, approximately 23% of the SGD group responded that the theory is a detailed explanation of the law or phenomena; however, it is based on law.In the SGD group, about 25% of respondents perceived that the theory is real/fact; however, this was discovered in a scientific study.Similarly, 40% of the conventional group responded that the theory is the idea of a researcher or an individual's view of the problem/phenomena.Even though they presented it in different ways, all groups had a naive view of scientific theory; therefore, all groups were similar in their views about the theory.

Social and cultural embeddedness in science
Under this criterion, the findings indicate that in all groups, more than 50% of respondents thought that culture and society did not affect scientific findings.The reason they provided for this claim is that the findings of science are applicable all over the world in the same way (the examples they provided were technological applications, gravitational acceleration, and measuring devices that are independent of culture).This indicates that all groups had naive views about social and cultural embeddedness in scientific investigations; thus, all groups had similar naive views about social and cultural embeddedness in science.
Under this criterion, the findings from the pre data indicate that more than 63% of the respondents in the SCGD, SSGD, and SGD groups said that scientists used their imagination and creativity to find new things.Their reason for this claim is that they believe that imagination and creativity are gifts for human beings.Therefore, scientists have used them to understand nature because it is an incomplete form.However, the conventional group opposed this idea and thought that scientists do not use their imagination and creativity; rather, they describe naturally existing things.The majority of the treatment groups had a better informed view about using creativity and imagination in the scientific investigation than the control group.

Using universal and/or different methods in scientific research
Under this criterion, the finding indicates that more than 61% of the respondents in each group thought that scientists use different methods.However, more than 16% of the respondents in each group indicated that a scientist used single and universal step-by-step methods.Respondents from all groups had relatively informed views about the methods used in scientific investigation, but still, in all groups, there were naïve views about the methods utilized in the scientific investigations.In this respect, the groups appeared to be similar.

Practicing of PS and NOS in physics laboratories
Under this criterion, the findings indicate that more than 63% of respondents in each group believed that they practiced PS and learned NOS in the physics laboratory better than in a formal course of study.Although there were similarities to some degree, there were also differences among the groups.Hence, about 40% of the respondents focused on the content knowledge they learned in lab work rather than using other process skills and/or concepts.

Models (processes) used in scientific research
In this criterion, the results indicate that all groups differ in their opinions about the use of different models in scientific research.About 58% of the SCGD and 60% of the conventional groups selected "B."However, 25%, 27%, 64%, and 20% of the SCGD, SSGD, SGGD, and conventional groups, respectively, were unsure.A small number of students in each group selected Choice "A." Choice B is a better scientific approach than the choice "A" (see the choices in Table 4).

What more to learn in science/physics laboratory
Under this criterion, the findings indicate that each group reflected different learning experiences in the physics/science lab (what to do and what to learn in the physics lab).Some of the groups focused on pedagogy, others on forms of laboratory, and others on the nature of science and process skills.Thus, in this respect, each group is different.

Pedagogical and forms of laboratory orientation
The findings revealed that all groups had different pedagogical orientations.This might be related to differences in their learning exposure (i.e., they learn from different instructors and teachers in schools and colleges, they have different experiences, and their selection of forms of laboratory and pedagogical orientation were different).

Motivation
Responses to questions concerning motivation indicated that each group was motivated differently towards laboratory work.As they learn from different teachers and through coursework in schools and colleges, they develop different motivations.Thus, all groups had different motivations towards physics laboratory work in physics.

Post conception about views of nature of science, pedagogy and forms of laboratory orientation, practicing PS, and motivation
After the intervention in the study design, the same tools were administered with minor modifications to investigate students' views on the nature of science, process skills, pedagogy, forms of laboratory orientation, and motivation.The themes obtained from the semi-structured questions are organized and presented in relation to the criteria used to measure the variables in the following section.The data presented under appendix Table A3.

Observation and interpretation in science
Under this criterion, the finding indicates that in the three treatment groups, on average, greater than 55% of the respondents thought that there was a difference in observation and interpretation of scientists for the same observation.The evidence was related to the scientists' backgrounds and differences in data collection techniques.However, only 30% of the participants in the conventional group supported this claim.Thus, groups had different views about observation and interpretation in science after the intervention of the study design.Regarding this idea, in the pre-test, the groups were similar, but in the post-administration period, the groups were different, and some students changed their views.This may be because of the impact of the models implemented in each group.

Changeability of theory and law in science
Under this criterion, the finding indicates that more than 50% of respondents from the SCGD, SGD, and control groups responded that theory and law were changeable.The evidence they provided for the claim was that when laws and theories were tested by other scientists, stronger and better evidence errors could be obtained.As a result, theories and laws can change.The same idea was supported by 36.36% of respondents from the SSGD group.In contrast, about 50% of the conventional, 45% of the SSGD, and 23% of the SGD groups perceived that theory and law are not changeable.The evidence they provided for this claim was that scientific laws and/or theory are based on facts and sustainable evidence; otherwise, it is not scientific.Thus, the groups had different views after the intervention of the study design.Regarding this idea, in the pre-test, the groups were similar, but in the post-administration period, the groups were different, and some students changed their views.This may be because of the effects of the study design implemented in each group.

Theory and law
Under this criterion, the findings indicate that 41.67% of SCGD, 63.63% of SGD, and 40% of the conventional groups perceived law as the steps used to conduct a scientific activity formulated by scientists.They perceived law as the culture of a certain scientific community or society.However, only 15.38% of respondents in the SGD-implemented group supported this view.More than 30% of the respondents from each group perceived that law is the finding of research or a proven idea, which is a constant or generalized idea that could be achieved after the agreement of scientists.Moreover, they believed that law is a mathematical model (formula) in any field of study.This indicates that after the intervention, all groups demonstrated naive views about scientific law.Thus, all groups were similar because the majority of students had a naïve view of laws in science/ physics.Regarding this idea, students had the same naïve view before and after the intervention.The implemented design had a smaller impact on the view.

Changeability of theory and law
From the data, it was also found that 58.33% of SCGD, 72% of SSGD, 38% of SGD, and 30% of respondents in each group perceived that the theory is a testable and changeable idea of the scientist.However, all groups had naive views on the theory.About 25% of SCGD groups perceived theory as a fact, and 25% of SCGD and 15% of SGD groups perceived theory as an explanation of the law.Even though there were changes after the interventions in all groups, they also demonstrated mixed views about scientific theory.Thus, all the groups had different views about the theory after the intervention of research design and, to some extent, the implemented study design had some impact.

Social and cultural embeddedness in scientific investigation
Under this criterion, the findings indicate that the majority of students from all groups (more than 53%) responded that culture and society do not affect scientific findings.The evidence they provided for their claim was that the findings of science are applicable all over the world in the same way.This naive view was equally supported by the three treatments and conventional groups; thus, the groups had a similar naive view about the social and cultural impact of scientific investigation.This indicates that the study design had less impact on this idea.

Using imagination and creativity in scientific investigation
Under this criterion, the findings revealed that more than 72% of the groups perceived that scientists used their imagination and creativity to find new things.The evidence they supplied to this idea is that imagination and creativity are gifts of human beings.Some respondents believed that nature does not have a structured and complete form for human needs.According to their responses, scientists need to use their imagination and creative abilities to adjust nature in a suitable form for human beings.This indicates that the majority of the students had informed views about using creativity and imagination in scientific investigation after the intervention.The implemented study design had a positive impact on the students' views.

The scientific methods used in scientific investigation
For this criterion, the finding indicates that 53% of SGD, 72% of SSGD, and 80% of conventional group respondents believed that scientists used a variety of methods in a scientific investigation.However, the majority of the SCGD group respondents believed that scientists used single and universal methods in scientific investigations.After the intervention, the SCGD group tended to change their responses based on the belief that scientists use varied methods of scientific investigation to single and/or universal methods.In an overall comparison of groups' views on using different and single methods in scientific investigation, the majority of respondents from all groups had informed views about the methods used in scientific investigation, and all groups had similar informed views.This indicates that, to a certain extent, the implemented study design had a positive impact on this view.

Practices of process skills and learning about NOS in physics laboratory
For this criterion from all groups, more than 70% of the respondents favored the practice of process skills in a better way than the formal class, and learned the nature of science in the physics laboratory after the intervention.Thus, all groups had a similar view about this issue, and the implemented study design had a positive impact on this view.

Models of scientific method/process
For this criterion, more than 60% of the SCGD and conventional group respondents selected steps (order) of the scientific process under choice B (see Tables 2 and 4), which was a better approach for them.However, more than 45% of the SSGD and SGD groups supported steps in choice A. However, after the intervention, groups had different views on using scientific process models.Moreover, after the intervention, none of the groups provided an alternative model.In addition, after the intervention, students in each group had naïve views about the models of scientific investigation, and the implemented study design had less impact on this idea.This may lead to a modification of the study design or an increase in the intensity of the intervention.

What more to learn in science/physics laboratory
The findings from the questions related to this concept indicate that each group reflected different views.Some focused on the pedagogical approach, others on content knowledge, and others on the manipulation of equipment.Hence, after the intervention, all groups had a different orientation to what to do and what they learned more in physics laboratories.

Pedagogical and forms of laboratory orientation
The results of the responses after the implementation of the study design indicate that approximately 41.67% of respondents from SCGD and 36.36% of respondents from SSGD groups selected semi-structured guided discovery.Moreover, 53.84% of the respondents in the SGD group and 40% of the respondents in the conventional group selected the conventional method.After the intervention, with the dominant vote, all groups detected the method of teaching and the forms of laboratory implemented in each group.This indicates that after the implementation of the study design, the teaching method implemented in each group was detected by the students.Thus, all groups after intervention differed in their pedagogical and laboratory orientation.This indicates that the implemented study design was detected by each group.

Motivation
The results indicated that the students were motivated by the semester's laboratory work project.This could be related to the use of modified guided discovery methods, addition of NOS and PS to the treatment groups, and addition of concepts related to the nature of science and process skills for the conventional groups.Thus, all the groups were similar in their motivation towards laboratory work during the semester in which the study was conducted.

Summary of qualitative results
In this section, based on pre-and post-administration semi-structured questions, the nature of the groups before and after intervention in terms of the dominant views about NOS, pedagogical and forms of laboratory orientation, practicing PS, and motivation.Table 2 illustrates the nature of the groups in terms of the aforementioned variables, before and after the intervention.A summary of the qualitative findings is presented in Table 2.
Table 2 shows the nature of the groups before and after the intervention, according to the criteria derived from the themes identified from the qualitative findings.This analysis aims to triangulate the findings from semi-structured data.The findings indicate that groups are similar in terms of NOS before and after the intervention, but somehow different in terms of pedagogical orientation, practice of process skills, and motivation-related questions.This implies that the implemented pedagogy/study design had an impact on pedagogical orientation, process skills, and motivation.However, it has less of an impact on the views of NOS-related concepts.

Among groups comparison in terms of covariates (post test)
Table 3 illustrates the group comparisons in terms of covariates after the intervention of the study design.
From the inferential statistical test, comparing the groups in terms of covariates used in this study indicated that after the intervention, none of the groups were statistically significant in terms of pedagogical orientation (p = 0.777), practicing level of process skills (p = 0.777), and academic performance (CGPA), p = .681.Moreover, the impact factor of the study design (Ƞ2) was small for all variables.The findings indicate that the explicit approach of PS, implicit approach of NOS, different types of pedagogy, and forms of laboratory did not make a significant difference among groups in pedagogical orientation, practicing PS, and overall academic performance (CGPA).

Comparison of groups in terms of dependent variables with control of covariates variables (after intervention)
Table 4 demonstrates the between-group comparison with control of covariates after the intervention of the study design.
The covariates appearing in the model were evaluated at the following values: pedagogy and forms of laboratory orientation = 16.3304,practice of process skills = 28.5982 and overall academic achievement (CGPA) = 2.6586.The results indicate that there is a significant difference among the groups in terms of the conceptual knowledge test (p = .013)and procedural knowledge test (p = .000), and motivation (p = .003); however, there was no significant difference in the views of NOS (p = .846).The adjusted Rsquared or the impact factor of implementing the study design on each variable was 0.097 (9.7%) for conceptual knowledge, 0.248 (24.8%) for procedural knowledge, 0.008 (0.8%) for NOS, and 0.122 (12.2%) for motivation.The results showed that the effect was small for all variables except procedural knowledge (Cohen, 1988).This suggests that the implicit approach of the nature of science with different laboratory setups, pedagogies, and explicit approaches to process skills have no more effect on students' views of the nature of science and did not make a significant difference among groups.However, the implemented study design showed significant differences among the groups in terms of conceptual and procedural knowledge and motivation.

Pair wise comparisons of groups in terms of dependent variables
Table 5 demonstrates the pair wise comparisons among the groups.
From pair wise comparison, as indicated in Table 5, conceptual knowledge between the control and three treatment groups was significant (p < 0.05), but there was no significant difference among treatment groups (p > 0.05).In addition, comparing groups in terms of procedural knowledge test, there was still a significant difference between the control and three treatment groups (p = 0.000), but there was no significant difference among the treatment groups.In addition, when comparing the groups in terms of NOS, all groups were not significant (P > 0.05).In addition, when comparing groups in terms of motivation, there was a significant difference between the control and SGD (p = 0.000), SGD and SSGD (p = 0.013), and SGD and SCGD (p = 0.019) groups.
This finding indicates that the study design (three modified guided discovery methods and control group) had a significant impact on and difference between the control group and modified guided discovery implemented groups in terms of conceptual and procedural knowledge and motivation.However, there was no difference among the groups in terms of views on the nature of science.In addition, the study design showed no significant difference among the modified guided discovery groups in conceptual and procedural knowledge and had a difference in terms of motivation.

Discussion
This study implemented a tandem design phase (III) using a quasi-experimental approach.This design was selected to compare the three paired matched treatment groups and one control group.In this study, a mixed method was used.Qualitatively, the study used empirical phenomenology and a quantitative descriptive comparative study.To triangulate the findings, a concurrent transformative approach using the QUAN-qual model was implemented.The results from qualitative explication indicate that before and after the intervention of the study design, all groups were similar in most areas of views of NOS and PS; however, after the intervention, groups differed in their pedagogical and forms of laboratory orientation, practicing of process skills, and motivation.The findings on exploring misconceptions about NOS and PS indicate that students had naïve views about theory, law, methods in scientific investigations, and the effect of culture and society on scientific investigations.This finding was corroborated by quantitative results.
Although there is no related literature that uses the same study design as indicated in Adisu et al. (2021b), the findings about the views of NOS and PS support the findings of Behiye et al. (2009).The effect of the study design on pedagogy and forms of laboratory orientation, practicing process skills, and overall academic performance indicates that there is no significant difference among groups after the intervention of study design, which is in agreement with the findings of Adisu and Abebaw (2021).Among group comparisons in terms of the dependent variable with control of covariates, there was a significant difference in conceptual and procedural knowledge and motivation, but not in the views of NOS and PS.The pair-wise comparison after intervention indicates that the conventional group was significantly different from the three modified guided-discovery methods in conceptual and procedural knowledge and motivation.This implies that the implemented study design (implicit activity-based approach of NOS and explicit approach of PS in different pedagogy and forms of laboratory) did not make any significant difference among groups in terms of views of NOS and PS.Although there are different pedagogies, forms of laboratory, and models of learning implemented in each group, the same implicit activity-based approach (answers to questions related to NOS and PS based on data gathered in the physics laboratory work) was implemented in each group.This means that the difference in pedagogy, forms of laboratory, and models of learning did not bring about a difference in students' views on NOS and PS.However, Baloy et al. (2017) and Clough (2007), using an explicit reflective question during intervention about NOS and PS under the guided-discovery method, found that there is a difference in students' views of NOS and PS, which contradicts the findings of this study.
One of the unique findings of this study is that the three levels of modified guided-discovery groups were not significant in conceptual and procedural knowledge, and all groups were similar in their views of NOS and PS.This implies that the model of learning used in the modified guideddiscovery groups had a similar impact on conceptual and procedural knowledge, NOS, and PS.However, the model had different effects on their motivation for laboratory work.There is no direct related literature in physics education laboratory work in the college of teacher education to oppose or support this finding that was conducted in the same way.
The overall result implies that implicit integration of NOS less promote students' advanced views towards different components of NOS and that explicit integration of PS less help them resolve naive conceptions about different levels of PS.However, the implemented learning model had a significant impact on students' motivation and different forms of scientific knowledge, that is, conceptual and procedural knowledge.This means that the model has helped improve conceptual and procedural understanding and promotes motivation towards conduct laboratory activities.
Furthermore, the findings of this study indicate that even though third-year pre-service physics teachers took more than three laboratory courses in physics, including this study design, the results showed that they held inadequate and naive conceptions of NOS and PS.This suggests the need for further studies in this area.This could also be related to the curriculum underuse in the college of teacher education that needs consideration in incorporating and integrating NOS and PS.Finally, the paucity of studies that were engaged in developing a variety of designs to investigate students' views, knowledge, and skills, and how to affect their learning outcomes limits the comparison of the current findings with those in the literature to some extent.This study did not investigate the impact of the implemented model on students' epistemic knowledge.In addition, the study did not include interviews with teachers or students.

Conclusion and educational implications
In this study, four learning models were implemented in physics laboratories.The models were represented in terms of three phases of modified guided-discovery methods: SGD, SSGD, and SCGD, and the conventional method.Then, the study developed, validated, and implemented a model lesson plan for the pedagogy of each model.The findings revealed that pre-and post-administration, there is a similarity between all groups in terms of NOS and PS, but different views about pedagogy and forms of laboratory orientation.It was found that modified guided discovery groups were significant in conceptual and procedural knowledge and motivation; however, they were similar in view of the nature of science.This indicates that the models used in each group (explicit approach of process skills and implicit approach of nature of science) did not make any difference among the groups.Thus, it is suggested that teachers use different models, such as modified guided-discovery methods, to help students promote conceptual and procedural knowledge and motivation.Furthermore, the physics laboratory curriculum should devise a way for laboratory materials to incorporate different learning models and be used in physics laboratories to enhance students' various forms of scientific knowledge.Finally, it is recommended that further studies investigate the impact of the design on epistemic knowledge and modify the study design to enhance students' informed views on NOS and PS in the physics laboratory.
Finally, this study revealed gaps and fills them in physics laboratory curricula of colleges' teacher education and literature in terms of integrating concepts, NOS and PS, forms of laboratory, pedagogy, model of learning, and their implementation.Therefore, this study revealed how learning models guide the selection, integration, and implementation of pedagogy and laboratory forms.Thus, this study has significant implications for science education laboratory work in colleges' and schools' teacher education.

Appendix Table A1. Alternative alignment of models of learning, pedagogies, forms of laboratory, NOS, and PS and contents in physics laboratory
The Thus, any type assessment tools can be used Adapted from Clough (2007), and reported as Daniel,et al. (2023) and Adisu et al. (2021a).

Figure 1 .
Figure 1.Procedure followed to conduct the study.
Most computing views identified about NOS and PS, pedagogical and forms of lab, practicing PS, and motivation (post/s of background knowledge difference and the observation techniques/tools they may use different

Table A3 . (Continued) Questions Most computing ideas identified Groups and frequency for each computing ideas SCGD (N = 12) SSGD (N = 11) SGD (N = 13) Conventional method(N = 10) Explication of questions related to Nature of science and process skills
generalized idea or proved idea after agreement of scientist, or it is mathematical model (formula in any field of study) • law is finding of research or proved idea, which is constant• law is

Table A3 . (Continued) Questions Most computing ideas identified Groups and frequency for each computing ideas SCGD (N = 12) SSGD (N = 11) SGD (N = 13) Conventional method(N = 10) Explication of questions related to Nature of science and process skills
, b/s without culture/society there is no science, it should be accepted before practice in society, so it is culture based findings are universal, and not affected by culture and society, Example gravitational acceleration, all measuring devices-etc their imagination and creativity, b/s science is the product of creativity((Example Ohm, Mari cure,-etc), hypothesis formulation needs imagination, creativity and imaginations are the practices/ nature of human being • yes affect• scientists independently conduct it from any culture, and scientists from many culture agreed before it distributed all over the world • -the • they use