Exploration of science teaching self-efficacy outside professional development context for inquiry-based teaching

Abstract

There are situations where teachers accustomed to years of traditional teaching are tasked with implementing inquiry-based science teaching in schools nationwide, with few professional developments to change their self-efficacy. The implicit expectations are that teachers’ self-efficacy will improve as they implement inquiry-based practices over time. This study explored self-efficacy levels of science instructors outside of the professional development context; and combined concepts from social cognitive and curriculum implementation theories and findings to explore effective ways of implementing inquiry-based teaching practices in schools. We collected, analysed, and triangulated quantitative and qualitative data from a larger sample of 308 and subsample of 18 junior high school science instructors. We found that science instructors outside the professional development context had low self-efficacy of inquiry-based but high self-efficacy of traditional teaching practices. We also found that there were statistically substantial differences in self-efficacy of science instructors outside of the professional development context, based on school type, school location, and academic qualification. From combination of concepts from social cognitive and curriculum implementation theories and findings from the study, it was ascertained that differentiated and phased approach to implementation of inquiry-based teaching practices needs to be adopted in various schools of diverse teachers with different self-efficacy levels. Different teachers in different schools with different levels of self-efficacy may start implementing inquiry-based practices with different content knowledge, teaching skills, and progress at different rates.

IMPACT STATEMENT

Inquiry-based science pedagogy is a constructivist approach that is effective for promoting meaningful learning and scientific literacy. Consequently, there have been science education reform initiatives to adopt and implement inquiry-based pedagogy in schools in many countries, including Ghana. However, in many situations, teachers with years of experiences in traditional instruction are those tasked with implementing the innovative approach, without adequate professional development to re-orient their self-efficacy. Findings from this study suggest that it is less likely to have significant and sustained improvements in instructors’ self-efficacy with few professional developments and rare engagements in inquiry practices over time. This study shows the importance of examining self-efficacy of science teachers in situations with no consistent opportunities for teachers to participate in professional development experiences. It also suggests the importance of adopting differentiated and phased approach to implementing inquiry-based practices in various schools with diverse teachers holding different levels of science teaching self-efficacy.

Keywords:

• Self-efficacy
• inquiry-based science teaching
• traditional science teaching
• junior high school
• professional development

REVIEWING EDITOR:

• Debra Laier Chapman, School of Computing, University of South Alabama, Mobile, Alabama, USA

SUBJECTS:

• Agriculture & Environmental Sciences
• Biology
• Earth Sciences
• Chemistry
• Physics
• Teachers & Teacher Education
• Theories of Learning
• Classroom Practice
• Middle School Education
• Secondary Education

Introduction

Reforms in science curricula that emphasize inquiry-based teaching and learning have been key educational issues globally for decades (e.g. National Research Council [NRC], 1996, 2000, 2012) and recurring themes in many educational systems (e.g. Curriculum Research and Development Division [CRDD], 2007, 2012; National Council for Curriculum & Assessment [NaCCA], 2020). High levels of self-efficacy are essential for successful and sustained implementation of inquiry-based science teaching in classrooms (e.g. Cetin-Dindar, 2022; Murphy et al., 2020; Perera et al., 2022).

There are situations where teachers accustomed to long years of traditional instruction are tasked with implementing inquiry-based teaching in schools nationwide, with few professional developments to change their self-efficacy. For instance, most of the teachers tasked with implementing inquiry-based pedagogy in basic schools in Ghana are individuals with initial teacher training and a substantial number of untrained and uncertified teachers, and are currently engaged in frequent traditional science teaching practices in classrooms (Buabeng et al., 2020; Mereku, 2000; Sofo et al., 2015; Tanaka, 2012). Although, it is mandatory for the teachers to participate regularly in School-Based INSETs (SBI), Cluster-Based INSETs (CBI), and other INSETs to improve their knowledge, skills, and experiences; and renew their licenses and teaching contracts (Japan International Cooperation Agency [JICA], 2014; National Teaching Council [NTC], 2020), the provision of INSETs have been inconsistent and circumstantial, and appears not to be specifically targeted at science teachers (Association for the Development of Education in Africa [ADEA], 2016).

Despite the inconsistent and circumstantial provision of professional development experiences, the expectations are that teachers’ capabilities and self-efficacy will improve as they implement inquiry-based practices over time. Therefore, examination of teachers’ self-efficacy is essential to ascertain whether they have the capability to implement inquiry pedagogy as envisioned in curricula documents. Numerous studies into science teaching self-efficacy conducted over the past decade focused on in-service teachers engaged in professional developments (e.g., Lee et al., 2022; Lotter et al., 2016; McKinnon & Lamberts, 2014; Murphy et al., 2020; Peters-Burton et al., 2015). There have been insufficient studies within the same period to explore science teachers’ self-efficacy outside the context of professional development for inquiry-based teaching. Given the large number of teachers with no consistent opportunities to participate in professional developments, it is important to examine teachers’ self-efficacy outside the contexts of professional developments, to determine the levels of their capabilities for inquiry-based teaching and take remedial action where necessary.

Again, only a few studies published over the past decade examined relationships between self-efficacy and years of teaching experience (Lotter et al., 2016; McKinnon & Lamberts, 2014) and gender (Lotter et al., 2016) of in-service teachers in professional development contexts. There have been insufficient studies within the same period to examine differences in self-efficacy of in-service teachers outside of the professional development context, based on school-type, school location, and academic qualification.

Additionally, most studies into science teaching self-efficacy were based on concepts from the social cognitive (Bandura, 1977) and social learning (Rotter, 1966) theories. To the best of our knowledge, no study has examined science teachers’ self-efficacy based on combination of concepts from the social cognitive and curriculum implementation theories. Based on the insufficient studies mentioned, we explored science teaching self-efficacy outside the context of professional development; and combined concepts from social cognitive and curriculum implementation theories and findings from the study to explore effective ways of implementing inquiry-based teaching in schools. The research questions used are as follows:

1. What are the self-efficacy levels of science instructors outside the professional development context?

2. Is there difference in the self-efficacy of science instructors outside the professional development context based on school-type?

3. Is there difference in the self-efficacy of science instructors outside the professional development context based on school location?

4. Is there difference in the self-efficacy of science instructors outside the professional development context based on academic qualification?

Conceptual framework

Inquiry-based science teaching and learning

Inquiry-based science teaching and learning is a novel student-centred approach that actively engages students in important, meaningful, and exciting hands-on activities to explore scientific phenomena in the physical environment. The inquiry approach engages students in procedural, epistemic, social, and conceptual activities under the guidance of teachers (Furtak et al., 2012). During procedural activities, students engage in and learn how to pose scientifically oriented questions, plan investigations, conduct investigations, and collect and record data (e.g. NRC, 1996, 2000, 2012). During epistemic activities, students engage in and learn how to examine and evaluate scientific data, resolve inconsistencies in data, interpret data, and learn that scientific knowledge is tentative (Crawford, 2000; Furtak et al., 2012). During conceptual activities, students engage in and learn how to apply prior scientific knowledge to interpret data and to check the validity of their interpretations. Again, students learn concepts contextualized in science processes. Inquiry-based social activities involve students engaging in and learning collaborative and communicative processes used to generate scientific knowledge (Gillies, 2008). In guidance activities, teachers facilitate, coach, mentor, lead, and direct student investigations (Crawford, 2000). Teachers use questions and scaffolds to guide student investigations (Lazonder & Harmsen, 2016). Teachers give autonomy to students to take responsibility for their learning.

Science teaching self-efficacy

Science teaching self-efficacy is a teacher’s belief in their capability to employ instructional strategies to improve students’ learning outcomes (Bandura, 1977; Tschannen-Moran et al., 1998). The two distinct but related components of science teaching self-efficacy are personal science teaching efficacy (PSTE) and science teaching outcome expectancy (STOE) (Bleicher, 2004; Enochs & Riggs, 1990; Riggs & Enochs, 1989). While PSTE is a teacher’s belief in their capabilities to employ appropriate strategies to teach science, STOE is the teacher’s belief that employment of these strategies will improve students’ achievements. Teachers must have both strong PSTE and STOE to increase the likelihood of implementing an instructional approach in classrooms.

Bandura (1977) noted that teachers may have different magnitudes, strengths, and generality of self-efficacy. While some teachers may hold strong general self-efficacy for innovative teaching tasks, others may hold weak specific self-efficacy for didactic instructional tasks. Strongly established traditional teaching self-efficacy and practices that have protected teachers for years cannot be easily discarded. When strongly established traditional self-efficacy and contextual teaching practices discount the importance of the inquiry-based approach, teachers may continue their old practices, thereby compromising the implementation of novel teaching initiatives (Bandura, 1977).

Teachers will approach, explore, and attempt inquiry practices that are within their perceived capabilities, but avoid practices that are beyond their perceived capabilities (Bandura, 1977). High self-efficacy is associated with an enhanced understanding of scientific processes, and willingness, enthusiasm, and confidence in employing inquiry practices in classrooms (Avery & Meyer, 2012). Frequent use and satisfaction with inquiry practices are directly linked to high self-efficacy (Perera et al., 2022). Teachers with high self-efficacy have higher perceptions of the constructivist learning environment, which emphasizes personal relevance, critical voice, and student engagement (Cetin-Dindar, 2022). Teachers with high self-efficacy are capable of designing inquiry projects that engage students in planning and implementing activities; collaborating and participating in problem-solving; analysing interconnections between environmental, economic, and societal issues; discussing and debating diverse issues; exploring and respecting diverse perspectives; and exploring and examining authentic problems and possible solutions (Murphy et al., 2020).

Influence of context and demographic factors on science teaching self-efficacy

Owing to the influence of contextual factors on how teachers cognitively process science instruction information, mastery experiences and efficacy expectations generated in particular circumstances do not produce strong generalizations to other situations (Bandura, 1977). While teachers may enthusiastically enact inquiry pedagogy in contexts where it is supported, they may be doubtful about initiating inquiry teaching in contexts where it is less encouraged. Similarly, teachers may demonstrate favourable self-efficacy and practices of the inquiry approach in contexts where it is supported but demonstrate didactic self-efficacy and practices in situations where it is not supported. Evidence shows that cohesive school support that promotes self-efficacy is one where administrators respond to teachers’ concerns and inspire teachers to attempt novel ideas and where teachers inspire their peers to respond to students’ issues and problems (Lotter et al., 2016). Teachers who lack support for science teaching in schools in which they work or who encounter challenges from students, administrators, and the educational system tend to have low self-efficacy. Discrepancies between teachers’ self-efficacy and teaching behaviours occur when there are misalignments between contextual factors and teaching demands (Bandura, 1977).

Evidence is inconclusive regarding the influence of demographic characteristics on the formation and effects of teachers’ self-efficacy. While some studies show that gender and years of teaching experience significantly influence science teachers’ self-efficacy (Lotter et al., 2016), other studies show no substantial gender differences in teachers’ self-efficacy (van Rooij et al., 2019), and no significant differences in PSTE and STOE based on gender and grade levels of preservice teachers (Gray, 2017).

Core constructs of curriculum implementation theory

The theory of curriculum implementation recognizes the existence of diversity and complexity of schools based on the diversity of funding opportunities, socioeconomic background, building infrastructure, teachers’ working experiences, students’ achievements, teaching and learning resources, and head teachers’ leadership styles among others (Rogan & Grayson, 2003). Therefore, failure to consider the diversity of teachers in different school contexts in implementation of innovative curricula will contribute to the poor outcomes of educational initiatives. Based on this, the theory emphasizes a differentiated and phased approach to the implementation of innovative teaching practices in schools nationwide (Rogan & Grayson, 2003).

Rogan and Grayson (2003) outlined three core constructs and six propositions in the curriculum implementation theory. The core constructs are, profile of implementation, capacity to support innovation, and support from outside agencies. Profile of implementation deals with establishing and disseminating a spectrum of teaching practices to initiate, adopt, and sustain the implementation of innovative curricula in schools (Rogan & Grayson, 2003). The profile of implementation allows curriculum planners, educators, and school administrators to determine their current status, identify their strengths, consider the school context and capacity, and select workable paths to implement innovative curricula. The capacity to support innovation involves efforts that must be made to understand and explain factors that enhance and/or impede the implementation of innovative curricula in schools. It recognizes that not all schools have a similar capacity to provide professional development and physical and human resources for innovative curriculum implementation (Rogan & Grayson, 2003). Outside agencies are organizations and institutions outside the school that cooperate with the school to implement innovative curricula (Rogan & Grayson, 2003). Outside support tends to provide funds, sponsorship, and expertise in curriculum implementation, particularly in developing education systems.

Methods

Research design

We employed the convergent parallel mixed methods design to collect numerical and qualitative data (Cohen et al., 2007; Creswell, 2013; Creswell & Plano Clark, 2007). We surveyed a large sample of integrated science instructors for numerical data collection and used multiple case studies of a subsample for qualitative data collection (Cohen et al., 2007). The quantitative and qualitative data were analysed separately and integrated during the presentation and discussion of the results. The convergent parallel mixed methods design allowed us to obtain complementary quantitative and qualitative data on the same phenomenon under investigation; combined the different strengths and non-overlapping weaknesses of quantitative and qualitative data and; directly compared and contrasted the quantitative and qualitative data to corroborate and validate the findings (Creswell, 2013; Creswell & Plano Clark, 2007). This design also allowed us to synthesize results from the two types of data to gain deep and complete insights into the issues investigated. A major limitation of the research design is the difficulty in resolving inconsistencies that may arise from quantitative and qualitative data.

Context of the study

The framework that guides professional development activities of public and private school teachers in the current study context stipulates that teachers must be engaged in mandatory, ranked, and recommended professional development activities to improve their knowledge, skills, and experiences and; renew their licenses and teaching contracts (National Teaching Council [NTC], 2020). This includes regular participation of teachers in School-Based INSETs (SBI), Cluster-Based INSTEs (CBI), District-Based INSETs (DBI), and other in-service trainings (Association for the Development of Education in Africa [ADEA], 2016; Japan International Cooperation Agency [JICA], 2014). According to this framework, renewal of teachers’ professional licenses and job contracts, promotion to higher ranks, and eligibility to hold certain administrative positions can be considered only when the teachers have been properly assessed and found to have earned satisfactory scores in their participation and performance in a number of mandatory, ranked, and recommended professional development activities (National Teaching Council [NTC], 2020).

However, the population of instructors from whom we drew our sample was not under any professional development programme to enhance their capabilities and self-efficacy in inquiry pedagogy at the time of our study. They were instructors outside the context of professional development. They were integrated science instructors engaged in routine day-to-day classroom teaching, assessment, and other curriculum activities in junior high schools (JHS) of various contextual factors. They were public and private school instructors in two rural districts, an urban municipality, and a rural-urban municipality in one coastal region of Ghana.

Sample

We used convenience, purposive, and stratified random sampling in a multistage process to select participants for the study. In the first stage, we conveniently and purposefully sampled two rural districts, one urban and one rural-urban municipalities in one coastal region of Ghana. The districts and municipalities were sampled because they were geographically not too far from the university where the researchers lived and; have schools in rural and urban contexts with different characteristics. Other districts and municipalities in the coastal region, as well as districts and municipalities in the northern, central, and other southern regions of the country were not sampled. We purposefully selected a larger sample of 308 science instructors from private and public JHSs in rural and urban localities in the selected region for quantitative data collection. In the third stage, we used stratified random sampling to select two public and two private JHSs from each district and municipality (a total of 16 JHSs) for multiple case studies. A total of 16 purposeful sample and two volunteer sample of science instructors from the schools were sub-selected for qualitative data collection.

School location (rural and urban) was one inclusion criterion for sampling the instructors. We used school location as an inclusion criterion because of differences in socio-economic, contextual, and technological circumstances of schools and instructors in rural and urban localities. Compared to rural schools, urban schools have access to additional resources for teaching technology, government financial support and business sponsorships, better human resource for planning and implementing teaching technology, and teachers with more familiarity and confidence in using teaching technology (Wang, 2013). Relative to rural schools, urban schools attract and retain more qualified science teachers, have better infrastructure and science equipment, and better management and supervision of teaching and learning activities (Addy, 2013; Somuah & Mensah, 2013).

School type (private and public) is another inclusion criterion used for sampling the science instructors. Differences in school types occur mainly from differences in teaching and learning practices (Mohammed et al., 2020), availability of resources, and assessment and administration practices. Compared to private schools, public schools tend to attract and retain more qualified and experienced science teachers (Chughati & Perveen, 2013). However, private schools tend to have fewer pupil to teacher ratio (Amoah et al., 2018), and better supervision of teaching and learning activities (Owusu et al., 2018).

Most of the sampled instructors were males 277(89.9%) and only a few were females 31(10.07%) (Figure 1a); most were public school instructors 176(57.14%) with 132(42.86%) being private school instructors (Figure 1b); 163(52.92%) instructors were from urban schools and 145(47.08%) from rural schools (Figure 1c); most were pre-university certificate instructors 200(64.94%), 105(34.09%) were graduate instructors, and a few were post-graduate instructors 3(0.97%) (Figure 1d); most of the instructors had 1–5 years teaching experience 178(57.79%), 79(25.65%) had 6–10 years teaching experience, 44(14.29%) had 11–15 years teaching experience, and a few had over 15 years teaching experience 7(2.27%) (Figure 1e).

Figure 1. (a) Distribution of science instructors by gender. (b) Distribution of science instructors by school type. (c) Distribution of science instructors by school location. (d) Distribution of science instructors by academic qualification. (e) Distribution of science instructors by years of teaching experience.

Instrument

A questionnaire and semi-structured interview schedule were used to collect data.

Teachers’ questionnaire

The questionnaire was designed to measure teachers’ self-efficacy in science teaching. Previous instruments (Bleicher, 2004; Riggs & Enochs, 1989; Smolleck et al., 2006) were used to guide the design and development of this instrument. The instrument was designed to measure teachers’ PSTE and STOE. Initially, 30 closed-ended questions were constructed. Responses to each question ranged from strongly disagree = 1, disagree = 2, uncertain = 3, agree = 4, to strongly agree = 5. Literature and expert judgements of two science educators were used to establish the content validity of the questions. The experts agreed on 18 items for pilot testing. We removed 12 items that were redundant. Three equivalent forms of the instrument were pilot-tested on 39, 52, and 108 samples of integrated science instructors. The preliminary item analysis showed that the first equivalent form exhibited the highest total reliability (α = .77). After deleting four items with low corrected item-total correlations, the total reliability (α = .78) of the first equivalent form increased and was adopted for data collection in the main study. After the main data collection, we conducted a confirmatory principal component analysis (PCA) with varimax rotation. After removing two items that did not load onto any component, and another two items with low corrected item-total correlations, the 10 remaining items that yielded two explainable components were isolated for analyses to answer research questions one, two, three, and four. The total variance (50.73%) explained by the two components was consistent with the results of similar studies (e.g. Bleicher, 2004; Lardy & Mason, 2011; Smolleck et al., 2006). The total reliability (α = .66) of the instrument and reliabilities of the components (PSTE, α = .79; STOE, α = .72) were acceptable for research.

Interview schedule

Open-ended items were constructed to collect the qualitative data. The items were used to elicit participants’ perceptions of PSTE and STOE. The open-ended questions enabled interviewees to provide thorough responses. It also enabled the researcher to ask probing and follow-up questions for additional and elaborate responses. Again, the open-ended questions enabled the researcher to ask questions about pertinent matters that evolved from the conversations (Jacob & Furgerson, 2012). It also permitted the researcher to vary the questioning order based on participants’ answers.

Data collection procedure

Questionnaires were administered to a larger sample of instructors to collect the numerical data. The return rate was 87.5% (308 fully completed questionnaires were retrieved from 352 questionnaires distributed). Physical collection of the quantitative data in first phase of the study lasted within 4 months. Again, we engaged a subsample of instructors in one-on-one face-to-face semi-structured interviews to collect qualitative data. Arrangements were made with the instructors to conduct the interviews. The first author conducted and audio-recorded all the interviews. The participants consented to participate in the study. Ethical issues explained on the consent forms included anonymity, confidentiality, privacy, and rights of instructors to decline participation or withdraw from the research. The majority of the interviews were completed within 45 minutes. It took about 2 months to finish the qualitative data collection in second phase of the study.

Data analysis

We screened the numerical data to retain fully completed questionnaires and removed extreme values. We reverse-coded the negatively worded questions. Checks on the assumptions of parametric analysis showed that the numerical data were normally distributed, exhibited linearity and homogeneity of variances and covariances, and had no multicollinearity.

Descriptive statistics were calculated from the numerical data. We used benchmark values outlined by van Aalderen-Smeets and van der Molen (2015) to classify the means as high, moderate, or low. Means equal to or less than 2 represented low (traditional-oriented) self-efficacy, means between 2 and 4 represented moderate self-efficacy, and means equal to or greater than 4 represented high (inquiry-oriented) self-efficacy. We calculated one-way MANOVAs using school type, school location, and academic qualification as categorical variables. PSTE and STOE were dependent variables. We also calculated means and standard deviations of items on the PSTE and STOE scales.

In the qualitative data analysis, the audio recorded interviews were transcribed, audited, and edited. The transcripts were imported into QSR Nvivo (version 10) for coding. Each transcript was read multiple times to identify codes that were relevant to the research questions. The iterative and inductive process allowed consistent application of codes that emerged from the data. The emergent codes were categorised into themes and patterns while relevant quotations were recorded. Results from the qualitative data analysis were directly compared and contrasted with results from the descriptive and inferential statistics to corroborate and validate findings from the study. Again, results from the descriptive and inferential statistics and qualitative analyses were synthesized to obtain deep and complete understanding of the problem investigated.

Results

Research question 1: What are the self-efficacy levels of science instructors outside the professional development context?

Using benchmark values outlined by van Aalderen-Smeets and van der Molen (2015) the quantitative results show that instructors outside of the professional development context had modest self-efficacy of science teaching (M = 3.23, SD = .51). Specifically, the instructors had both modest STOE (M = 2.79, SD = .69) and PSTE (M = 3.89, SD = .75) of science teaching (Table 1).

Table 1. Descriptive statistics of instructors’ personal science teaching efficacy (PSTE) and science teaching outcome expectancy (STOE).

Component of self-efficacy	Responses
	SD	D	U	A	SA	Average item mean	Average item standard deviation
	n(%)	n(%)	n(%)	n(%)	n(%)
Personal science teaching efficacy (PSTE)						3.89	.75
I wonder if I have the necessary skills to teach science content embedded in the processes of science.	81(26.3)	142(46.1)	47(15.3)	29(9.4)	9(2.9)
It is difficult to guide JHS students to evaluate and examine the quality of data.	85(27.6)	177(57.5)	22(7.1)	24(7.8)	0(0.0)
I don’t know what to do to turn JHS students on to do science like actual scientists.	65(21.1)	153(49.7)	43(14.0)	41(13.3)	6(1.9)
When a JHS student has a difficulty in how to manipulate science equipment, I am usually at a loss as to how to help him do it.	93(30.2)	148(48.1)	23(7.5)	44(14.3)	0(0.0)
Science teaching outcome expectancy (STOE)						2.79	.69
Increased effort in writing science concepts on the board for students to copy is related to their performance.	21(6.8)	48(15.6)	70(22.7)	135(43.8)	34(11.0)
If parents comment that their child is showing more interest in science, it is probably due to the teacher telling students answers they are expected to learn.	23(7.5)	73(23.7)	85(27.6)	77(25.0)	50(16.2)
If JHS students are not performing well in collecting and recording science data, it is likely due to ineffective teaching of how to collect and record data.	34(11.0)	47(15.3)	103(33.4)	98(31.8)	26(8.4)
The teacher is generally responsible for the achievements of JHS students in how to formulate hypothesis of science phenomena.	8(2.6)	55(17.9)	114(37.0)	102(33.1)	29(9.4)
If JHS students are not performing like actual scientists do, it is likely due to ineffective teaching of how scientists work.	34(11.0)	94(30.5)	83(26.9)	79(25.6)	18(5.8)
JHS students’ achievements in science is directly related to their teacher’s effectiveness in defining and explaining science concepts and principles.	0(0.0)	33(10.7)	34(11.0)	161(52.3)	80(26.0)
Self-efficacy of science teaching						3.23	.51
Note: High mean is inquiry-based, low mean is traditional-oriented							N = 308

SD = Strongly disagree, D = Disagree, U = Uncertain, A = Agree, SA = Strongly agree.

Most of the instructors expressed disagreements in outcome expectations of inquiry-based but agreements in outcome expectations of traditional science teaching practices. More than half of the instructors 161(52.3%) agreed and many other instructors 80(26.0%) strongly agreed that students’ achievements in science are directly related to the teacher’s effectiveness in traditional teaching practice of defining and explaining science concepts and principles. Again, many instructors 135(43.8%) agreed and another 34(11.0%) instructors strongly agreed that increased effort in traditional teaching practice of writing science concepts on the board for students to copy is related to the students’ performance (Table 1).

By contrast, many instructors 47(15.3%) disagreed, 34(11.0%) instructors strongly disagreed, and many more instructors 103(33.4%) were uncertain that inquiry-based teaching practices are effective in enabling students to collect and record scientific data. Similarly, many instructors 55(17.9%) disagreed, other instructors 8(2.6%) strongly disagreed, and many more instructors 114(37.0%) were uncertain that inquiry-based teaching practices are effective in enabling students to formulate hypotheses for science phenomena. Likewise, 94(30.5%) instructors disagreed, 34(11.0%) instructors strongly disagreed, and another 83(26.9%) instructors were uncertain about the effectiveness of inquiry-based teaching practices to enable students work like scientists (Table 1).

Less than half of the instructors agreed that inquiry-based teaching practices are effective in enabling students to collect and record scientific data 98(31.8%), formulate hypotheses for science phenomena 102(33.1%), and work like scientists 79(25.6%).

Results from the qualitative interviews also indicate that most instructors outside of the professional development context had low STOE in inquiry-based but high STOE in traditional teaching practices. The interviewees expressed disbeliefs in the outcome expectations of inquiry-based teaching practices. They said that JHS students are more familiar with rote learning and engaging them in inquiry-based teaching would be unproductive.

When JHS students are allowed to engage in inquiry, their performance will be abysmal. The students started rote learning from kindergarten … what they know is rote learning …. (Science teacher 12). Table 1 shows that most instructors outside of the professional development context expressed less disagreements in their capabilities in inquiry-based teaching practices. Many instructors 142(46.1%) disagreed and many other instructors 81(26.3%) strongly disagreed that they wonder if they have the necessary skills to teach content embedded in the processes of science. Similarly, more than half of the instructors 177(57.5%) disagreed and another 85(27.6%) instructors strongly disagreed that it is difficult for them to guide learners to examine and evaluate the quality of scientific data. Likewise, nearly half of the instructors 153(49.7%) disagreed and many other instructors 65(21.1%) strongly disagreed that they don’t know what to do to turn learners on to do science like scientists. Additionally, nearly half of the instructors 148(48.1%) disagreed and another 93(30.2%) instructors strongly disagreed that when a student has a difficulty in how to manipulate science equipment and materials, they are usually at a loss as to how to help the student do it better.

However, results from the qualitative interviews indicate that the PSTE of instructors outside of the professional development context might not be as high as the quantitative results appear to show. Most of the instructors had high PSTE based on traditional understandings of teaching, while others had high PSTE based on partial conceptions of inquiry. Instructors with high PSTE based on partial conceptions of inquiry expressed confidence in their capabilities to implement some inquiry-based and traditional teaching practices. They believed that they could effectively implement inquiry practices such as child-centred science lessons; active involvement of students in manipulating materials; guiding students to understand the processes in arriving at solutions to problems; and guiding students to formulate hypotheses for science phenomena.

The lesson should be planned in a way that is very interesting. Based on what the students already know, the teacher helps them to predict what they are going to get from their investigations. (Science teacher 2)

The planning of science teaching should be focused on the children. Everything should be centred on them because … they are the target of science lessons. (Science teachers 9)

Additionally, they believed that they could effectively implement traditional teaching practices, including giving students step-by-step procedures to conduct hands-on activities.

Say, if we are going to test for starch, you [the teacher] give them [the students] the first step … when they finish you give them the next step and continue until the activity is finished. So, I only give the order ‘do this, do that’. (Science teacher 18)

In contrast, instructors with high PSTE based on didactic understandings of teaching mostly expressed confidence in their capabilities to implement traditional practices. They believed that they could effectively teach science through traditional methods such as teaching content separately in one lesson, followed by hands-on activities in another lesson to verify the content already taught; using step-by-step processes and safety precautions for students to try out hands-on activities; and teacher trying out experiments prior to lessons to assess whether the materials are working properly.

After going through the theoretical aspect with students then you organise the students to go to the lab for hands-on activities during the next period. (Science teacher 3)

The teacher can also perform the practical in advance to find out if the materials are workable indeed. (Science teacher 16)

Research question 2: is there difference in the self-efficacy of science instructors outside the professional development context based on school-type?

One-way MANOVA results show that there was a statistically substantial school-type difference in self-efficacy of science instructors outside the professional development context, Wilk’s λ = .941, F(2, 305) = 9.482, p<.05, partial ŋ² = .059 (Table 2).

Table 2. One-way MANOVA of differences in science instructors’ self-efficacy based on school-type.

Effect	Value	F	Hypothesis df	Error df	Sig	Partial ŋ^2
Pillaice trace	.059	9.482*	2	305	.000	.059
Wilk’s Lambda	.941	9.482*	2	305	.000	.059
Hotelling’s trace	.062	9.482*	2	305	.000	.059
Roy’s largest root	.062	9.482*	2	305	.000	.059

* p < .05, N = 176(Public JHS instructors), N = 132(Private JHS instructors).

Subsequent one-way ANOVAs show that there were statistically significant school-type differences in ’ STOE, F(1, 306) = 6.045, p<.025 (Bonferroni adjusted alpha level), partial ŋ² = .019; and PSTE, F(1, 306) = 12.367, p<.025 (Bonferroni adjusted alpha level), partial ŋ² = .039 of instructors outside the professional development context (Table 3).

Table 3. One-way ANOVAs of differences in instructors’ STOE and PSTE based on school-type.

Component of self-efficacy	Source of variation	Sum of squares	df	Mean square	F	Sig	Partial ŋ^2
STOE	Between groups	101.007	1	101.007	6.045*	.015	.019
	Within groups	5113.214	306	16.710
	Total	91728.000	308
PSTE	Between groups	106.027	1	106.027	12.367*	.001	.039
	Within groups	2623.402	306	8.573
	Total	77410.000	308

* p<.025, N = 176(Public JHS instructors), N = 132(Private JHS instructors).

Private JHS instructors outside of the professional development context (M = 2.68, SD = .66) had lower STOE in inquiry-based but higher STOE in didactic-oriented teaching practices than public JHS instructors outside the professional development context (M = 2.88, SD = .69). Compared to public school instructors, private school instructors expressed more agreements in outcome expectations of didactic teaching practices. They expressed more agreements that students’ interest, achievements, and academic performance in science are directly related to the effectiveness of the teacher in traditional practices of defining, explaining, and writing science concepts on the board; and telling answers that students are expected to learn (Table 4).

Table 4. Average item means and average item standard deviations of public and private JHS instructors’ ratings of items on science teaching outcome expectancy (STOE).

Item	Average item mean		Average item standard deviation
	Public JHS teachers	Private JHS teachers	Public JHS teachers	Private JHS teachers
Increased effort in writing science concepts and principles on the board for JHS students is directly related to their academic performance.	2.79	2.42	1.14	.98
If parents comment that their child is showing interest in science at school, it is probably due to the teacher telling students answers they are expected to learn.	2.98	2.59	1.10	1.26
If students are not performing well in collecting and recording scientific data, it is most likely due to ineffective teaching of how to collect and record data.	3.17	3.05	1.18	1.03
The teacher is generally responsible for the achievements of students in how they formulate hypotheses for science phenomena.	3.30	3.28	1.02	.87
If JHS students are not performing like actual scientists do, it is most likely due to ineffective teaching of how actual scientists work.	2.80	2.91	1.15	1.05
JHS students’ achievements in science are directly related to their teacher’s effectiveness in defining and explaining science concepts and principles	2.23	1.85	.90	.83
Science teaching outcome expectancy (STOE)	2.88	2.68	0.69	0.66

N = 176(Public JHS instructors), N = 132(Private JHS instructors).

Note: High means represent inquiry STOE, low means represent traditional STOE.

By contrast, compared to private school instructors, public school instructors expressed less uncertainty in outcome expectations of inquiry-based teaching practices. They were less uncertain that ineffective inquiry teaching practices are most likely to lead to students’ non-performance in collecting and recording data; and that students’ achievements in formulation of hypotheses for science phenomena are generally the responsibility of the teacher.

Again, the quantitative results show that public school instructors outside of the professional development context (M = 4.02, SD = .72) had higher PSTE in inquiry-based teaching than their private school counterparts (M = 3.72, SD = .75). Compared to private school instructors, public school instructors expressed less disagreements in their capabilities in inquiry teaching practices. They expressed less disagreements that they have difficulty guiding learners to examine and evaluate the quality of scientific data; are usually at a loss as to how to help a learner in difficulty of manipulating science equipment and materials to do it better; wonder if they have the necessary skills to teach content embedded in the processes of science; and don’t know what to do to turn students on to do science like scientists (Table 5).

Table 5. Average item means and average item standard deviations of public and private JHS instructors’ ratings of items on personal science teaching efficacy (PSTE).

Item	Average item mean		Average item standard deviation
	Public JHS teachers	Private JHS teachers	Public JHS teachers	Private JHS teachers
I wonder if I have the necessary skills to teach content embedded in the processes of science.	3.97	3.65	1.00	1.01
I find it difficult guiding JHS students to examine and evaluate the quality scientific data on their own.	4.15	3.91	.74	.88
I don’t know what to do to turn JHS students on to do science like actual scientists do.	3.90	3.54	.94	1.04
When a JHS student has difficulty in how to manipulate science equipment and materials, I am usually at a loss as to how to help him do it better.	4.05	3.80	.92	1.02
Personal science teaching efficacy (PSTE)	4.02	3.72	0.72	0.75

N = 176(Public JHS instructors), N = 132(Private JHS instructors).

Note: High means represent inquiry-based PSTE, low means represent traditional PSTE.

However, insights from the interview results indicate that, although public school instructors outside of the professional development context had relatively higher PSTE than their counterparts in private schools, PSTEs of both groups might not be as high as the quantitative results appear to show. Most public and private school instructors had high PSTE based on didactic understandings of teaching, while others had high PSTE based on partial understandings of inquiry. While more public school instructors expressed confidence in their capabilities to implement inquiry teaching practices than private school instructors, more private school instructors expressed confidence in their capabilities to implement traditional teaching practices. Compared to two public school instructors, five private school instructors expressed confidence in their capabilities in traditional teaching practices. They believed that they could effectively monitor learners to perform science experiments through the traditional practice of putting learners into groups and giving them rules and regulations to follow.

I can effectively monitor JHS students to perform experiments … I will make sure that the students understand what they have to do and what they don’t have to do. (Science teacher 7)

Additionally, two private school instructors believed that they could effectively teach science when it involved traditional strict supervision of students’ behaviours.

There should be strict supervision. You have to ensure that children comply with the time to go to the laboratory for experiment. (Science teacher 3)

No public school instructor made remark about such traditional teaching practices.

Research question 3: is there difference in the self-efficacy of science instructors outside the professional development context based on school location?

One-way MANOVA results show that there was a statistically substantial school location difference in self-efficacy of science instructors outside the professional development context, Wilk’s λ = .975, F(2, 305) = 3.942, p<.05, partial ŋ² = .025 (Table 6).

Table 6. One-way MANOVA of differences in science instructors’ self-efficacy based on school location.

Effect	Value	F	Hypothesis df	Error df	Sig	Partial ŋ^2
Pillaice trace	.025	3.942*	2	305	.020	.025
Wilk’s Lambda	.975	3.942*	2	305	.020	.025
Hotelling’s trace	.026	3.942*	2	305	.020	.025
Roy’s largest root	.026	3.942*	2	305	.020	.025

* p < .05, N = 145(Rural JHS instructors), N = 163(Urban JHS instructors).

Follow-up one-way ANOVAs show that there was no statistically substantial school location difference in instructors’ STOE outside the professional development context, F(1, 306) = 2.580, p>.025 (Bonferroni adjusted alpha level), partial ŋ² = .008; but there was a statistically substantial school location difference in the instructors’ PSTE outside the professional development context F(1, 306) = 5.197, p<.025 (Bonferroni adjusted alpha level), partial ŋ² = .017 (Table 7).

Table 7. One-way ANOVAs of differences in instructors’ STOE and PSTE based on school location.

Component of self-efficacy	Source of variation	Sum of squares	df	Mean square	F	Sig	Partial ŋ^2
STOE	Between groups	5.593	1	43.593	2.580	.109	.008
	Within groups	5170.628	306	16.897
	Total	91728.000	308
PSTE	Between groups	45.583	1	45.583	5.197*	.023	.017
	Within groups	2683.846	306	8.771
	Total	77410.000	308

* p<.025, N = 145(Rural JHS instructors), N = 163(Urban JHS instructors).

Compared to their urban school counterparts (M = 3.80, SD = .74), rural school instructors outside of the professional development context (M = 4.0, SD = .75) had higher PSTE in inquiry-based teaching (Table 8). They expressed less disagreements in their capabilities in inquiry teaching practices. They expressed less disagreements that they find it difficult guiding students to examine and evaluate the quality of scientific data; are usually at a loss as to how to help a learner in difficulty of manipulating science equipment and materials to do it better; wonder if they have the necessary skills to teach content embedded in the processes of science; and don’t know what to do to turn students on to do science like scientists.

Table 8. Average item means and average item standard deviations of rural and urban JHS instructors’ ratings of items on personal science teaching efficacy (PSTE).

Item	Mean		Standard deviation
	Rural JHS teachers	Urban JHS teachers	Rural JHS teachers	Urban JHS teachers
I wonder if I have the necessary skills to teach content embedded in the processes of science.	3.88	3.79	1.04	.99
I find it difficult guiding JHS students to examine and evaluate the quality scientific data on their own.	4.13	3.98	.74	.87
I don’t know what to do to turn JHS students on to do science like actual scientists do.	3.87	3.64	1.00	.99
When a JHS student has difficulty in how to manipulate science equipment and materials, I am usually at a loss as to how to help him do it better.	4.10	3.80	.93	.99
Personal science teaching efficacy (PSTE)	4.00	3.80	.75	.74

N = 145(Rural JHS instructors), N = 163(Urban JHS instructors).

Note: High means represent inquiry-based PSTE, low means represent traditional PSTE.

However, insights from the interview results indicate that, although rural JHS instructors outside the professional development context had relatively higher PSTE than their counterparts in urban JHSs, the PSTEs of both groups might not be as high as the quantitative results appear to show. Most of the rural and urban school instructors had high PSTEs based on didactic understandings of science teaching, while others had high PSTEs based on partial understandings of inquiry. While both rural and urban school instructors espoused similar PSTE in inquiry teaching practices, more urban school instructors espoused PSTE in traditional teaching practices than rural school instructors. Compared to one rural school instructor, three urban school instructors believed that they could effectively teach science when they use the traditional practice of putting students into groups and giving them experimental procedures to follow.

Experiments will be effective … if I have sizeable groups of students following step-by-step procedures to perform experiments. (Science teacher 6)

Research question 4: is there difference in the self-efficacy of science instructors outside the professional development context based on academic qualification?

One-way MANOVA results show that there was a statistically substantial academic qualification difference in self-efficacy of science instructors outside the professional development context, Wilk’s λ = .942, F(4, 608) = 4.648, p<.05, partial ŋ² = .030 (Table 9).

Table 9. One-way MANOVA of differences in science instructors’ self-efficacy based on academic qualification.

Effect	Value	F	Hypothesis df	Error df	Sig	Partial ŋ^2
Pillaice trace	.058	4.594*	4	610	.001	.029
Wilk’s Lambda	.942	4.648*	4	608	.001	.030
Hotelling’s trace	.062	4.702*	4	606	.001	.030
Roy’s largest root	.062	9.401*	2	305	.000	.058

* p < .05, N = 200(Pre-university certificate instructors), N = 105(Graduate instructors), N = 3(Postgraduate instructors).

Follow-up one-way ANOVAs show statistically substantial academic qualification differences in instructors’ PSTE, F(2, 305) = 3.698, p<.05, partial ŋ² = .024; and STOE, F(2, 305) = 5.448, p<.025 (Bonferroni adjusted alpha level), partial ŋ² = .034, outside the professional development context (Table 10).

Table 10. One-way ANOVAs of differences in instructors’ STOE and PSTE based on academic qualification.

Component of self-efficacy	Source of variation	Sum of squares	df	Mean square	F	Sig	Partial ŋ^2
STOE	Between groups	179.841	2	89.920	5.448*	.005	.034
	Within groups	5034.380	305	16.506
	Total	91728.000	308
PSTE	Between groups	64.616	2	32.308	3.698**	.026	.024
	Within groups	2664.812	305	8.737
	Total	77410.000	308

* p<.025, **p<.05, N = 200(Pre-university certificate instructors), N = 105(Graduate instructors), N = 3(Postgraduate instructors).

Further analysis shows that there was a statistically substantial pair-wise difference in PSTE between pre-university certificate and graduate instructors, but no statistically substantial pair-wise differences in PSTE between pre-university certificate and postgraduate instructors, and graduate and postgraduate instructors (Table 11).

Table 11. Scheffe post-hoc analysis of differences in instructors PSTE based on academic qualification.

Dependent variable	Qualification of respondent (I)	Qualification of respondent (J)	Mean difference (I – J)	Standard error	Sig
PSTE	Pre-university certificate	Graduate	−.9555*	.3562	.029
		Postgraduate	−1.0983	1.7193	.816
	Graduate	Pre-university certificate	.9555*	.3562	.029
		Postgraduate	−.1429	1.7308	.997
	Postgraduate	Pre-university certificate	1.0983	1.7193	.816
		Graduate	.1429	1.7308	.997

^* *p<.05, N = 200(Pre-university certificate instructors), N = 105(Graduate instructors), N = 3(Postgraduate instructors).

Graduate instructors outside the professional development context (M = 4.05, SD = .73) had higher PSTE in inquiry-based teaching than their counterparts with pre-university certificates (M = 3.81, SD = .74). Graduate instructors expressed less disagreements in their capabilities in inquiry-based teaching practices than pre-university certificate teachers. They expressed less disagreements that they find it difficult guiding learners to examine and evaluate the quality of scientific data; are usually at a loss as to how to help a learner in difficulty of manipulating science equipment and materials to do it better; wonder if they have the necessary skills to teach content embedded in science processes; and don’t know what to do to turn learners on to do science like scientists (Table 12).

Table 12. Average item means and average item standard deviations of graduate and certificate JHS instructors’ ratings of items on personal science teaching efficacy (PSTE).

Item	Mean		Standard deviation
	Graduate JHS teachers	Certificate JHS teachers	Graduate JHS teachers	Certificate JHS teachers
I wonder if I have the necessary skills to teach content embedded in the processes of science.	3.98	3.75	1.00	1.02
I find it difficult guiding JHS students to examine and evaluate the quality scientific data on their own.	4.15	4.00	.76	.83
I don’t know what to do to turn JHS students on to do science like actual scientists do.	3.96	3.64	.96	1.00
When a JHS student has difficulty in how to manipulate science equipment and materials, I am usually at a loss as to how to help him do it better.	4.10	3.85	.88	1.01
Personal science teaching efficacy (PSTE)	4.05	3.81	.73	.74

N = 200(Pre-university certificate JHS instructors), N = 105(Graduate JHS instructors).

Note: High means represent inquiry-based PSTE, low means represent traditional PSTE.

However, insights from the interviews results indicate that although graduate instructors expressed relatively higher PSTEs than pre-university certificate instructors, PSTEs of both groups might not be as high as the quantitative results appear to show. Most graduate and pre-university certificate instructors had high PSTEs based on traditional understandings of science teaching, while others had high PSTEs based on partial understandings of inquiry. While both graduate and pre-university certificate instructors expressed similar PSTE in inquiry-based teaching practices, more pre-university certificate instructors expressed PSTE in traditional teaching practices than graduate instructors. In contrast to remarks from seven graduate instructors, 15 pre-university certificate instructors remarked that they could effectively teach science through traditional practices of showing and describing organisms, events, and phenomena; and that they could engage students in hands-on activities only when they ensured discipline and quietness in the classroom.

Say, if you are teaching about flowers in class, you [the teacher] go out and pluck a flower. You then take off the parts of the flower one by one and describe them to the students. (Science teacher 11)

… the teacher needs not tolerate noise in the class … the teacher should ensure discipline in class all the time … ensuring discipline in class involves proper use of punishment … (Science teacher 17)

Again, post-hoc analysis shows there was a statistically significant pair-wise difference in STOE between pre-university certificate (M = 2.70, SD = .65) and graduate (M = 3.00, SD = .72) instructors. However, there were no statistically significant pair-wise differences in STOE between graduate and postgraduate instructors, and between pre-university certificate and postgraduate instructors (Table 13).

Table 13. Scheffe post-hoc analysis of differences in instructors STOE based on academic qualification.

Dependent variable	Qualification of respondent (I)	Qualification of respondent (J)	Mean difference (I – J)	Standard error	Sig
STOE	Pre-university certificate	Graduate	−1.5233*	.4896	.009
		Postgraduate	−3.1233	2.3632	.419
	Graduate	Pre-university certificate	1.5233*	.4896	.009
		Postgraduate	−1.6000	2.3789	.798
	Postgraduate	Pre-university certificate	3.1233	2.3632	.419
		Graduate	1.6000	2.3789	.798

^* *p<.05, N = 200(Pre-university certificate instructors), N = 105(Graduate instructors), N = 3(Postgraduate instructors).

Pre-university certificate instructors outside of the professional development context expressed lower outcome expectations in inquiry-based but higher outcome expectations in didactic-oriented teaching practices than graduate instructors outside of the professional development context. Pre-university certificate instructors expressed higher outcome expectations in didactic teaching practices, that students’ achievements, interest, and academic performance in science are directly related to the effectiveness of the teacher in traditional practices of defining, explaining, and writing science concepts and principles on the board; and telling answers students are expected to learn (Table 14).

Table 14. Average item means and average item standard deviations of graduate and pre-university certificate JHS instructors’ ratings of items on science teaching outcome expectancy (STOE).

Item	Mean		Standard deviation
	Graduate JHS instructors	Pre-university certificate JHS teachers	Graduate JHS instructors	Pre-university certificate JHS instructors
Increased effort in writing science concepts and principles on the board for JHS students is directly related to their academic performance.	2.89	2.49	1.15	1.03
If parents comment that their child is showing interest in science at school, it is probably due to the teacher telling students answers they are expected to learn.	3.15	2.62	1.10	1.18
If students are not performing well in collecting and recording scientific data it is most likely due to ineffective teaching of how to collect and record data.	3.21	3.06	1.20	1.07
The teacher is generally responsible for the achievements of students in how they formulate hypotheses for science phenomena.	3.37	3.25	1.11	.87
If JHS students are not performing like actual scientists do, it is most likely due to ineffective teaching of how actual scientists work.	3.00	2.97	1.73	1.21
JHS students’ achievements in science are directly related to their teacher’s effectiveness in defining and explaining science concepts and principles	2.33	2.14	1.53	.78
Science teaching outcome expectancy (STOE)	3.00	2.70	.72	.65

N = 200(Pre-university certificate JHS instructors), N = 105(Graduate JHS instructors).

Note: High means represent inquiry STOE, low means represent traditional STOE.

By contrast, graduate instructors expressed less uncertainty in the outcome expectations of inquiry-based teaching practices than pre-university certificate instructors. Graduate instructors expressed less uncertainty that if students are not performing well in collecting and recording scientific data, it is most likely due to ineffective inquiry teaching practices; that the teacher is generally responsible for the achievements of students in the formulation of hypotheses for science phenomena; and that if students are not working like scientists do, it is most likely due to ineffective inquiry-based teaching practices.

Discussion

Integration of the quantitative and qualitative results shows that instructors outside of the professional development context had low science teaching self-efficacy. Specifically, the instructors had both low personal science teaching efficacy (PSTE) and science teaching outcome expectancy (STOE) of inquiry-based pedagogy, but high personal science teaching efficacy and science teaching outcome expectancy of traditional pedagogy. Despite their engagements in classroom science teaching practices over time, either the expected improvements in self-efficacy of instructors outside the professional development context did not occur or any improvements that occurred were unsustainable. This finding is a significant contribution to teaching practice and the literature. This finding is contrary to findings from a similar study conducted outside the professional development context (Cetin-Dindar, 2022). In the study by Cetin-Dindar (2022), teachers outside the professional development context had high self-efficacy because they frequently engaged in inquiry practices in classrooms, where they connected science lessons to real-life experiences and created opportunities for students to interact and express their thoughts. Numerous studies conducted in many situations have shown that most teachers frequently engage in didactic teaching (Ampiah, 2008; Buabeng et al., 2014; Ngman-Wara, 2015) and rarely engage in inquiry practices (e.g. Mohammed et al., 2020). The current aforementioned finding suggests that improvements in self-efficacy of instructors outside the professional development context did not occur because either they mostly engaged in didactic teaching or rarely engaged in inquiry practices over time.

Social cognitive theory (Bandura, 1977) indicates that teachers with low self-efficacy are unlikely to approach, explore, or attempt inquiry practices that are beyond their perceived capabilities. Likewise, teachers with both low PSTE and STOE are likely to abandon inquiry practices easily when they encounter challenges (Bandura, 1977; Riggs & Enochs, 1989). With the low personal science teaching efficacy and science teaching outcome expectancy of instructors outside the professional development context, they are less likely to approach, explore, and attempt inquiry-based practices that are beyond their perceived capabilities. However, in the event that the instructors attempt to implement or adopt inquiry-based pedagogy, they are likely to abandon it easily when they encounter challenges. Therefore, implementation of inquiry-based science curricula with the population of instructors we studiedoutside the professional development context is likely to fail or produce minimal outcomes.

The current finding is contrary to most findings from the contexts of methods courses and professional developments (e.g., Avery & Meyer, 2012; Lee et al., 2022; Murphy et al., 2020; Naidoo & Naidoo, 2021; Perkins Coppola, 2019; Seung et al., 2019), where teachers experienced significant enhancements in their self-efficacy because they had opportunities to engage in, observe, discuss, and overcome their fears of inquiry pedagogy (e.g. Gray, 2017; Kruse et al., 2021; Menon, 2020; Menon & Sadler, 2018; Webb & LoFaro, 2020). Interviews with our study participants showed that they had had few engagements in previous professional developments; but those engagements were not specifically designed to enhance instructors’ capabilities for inquiry-based science teaching and learning.

As a science teacher I have attended in-service training once since completing the training college four years ago. [Science teacher 12]

The workshop was not only on science, it was science and mathematics workshop …. [Science teacher 3]

It is less likely for instructors to experience sustained improvements in inquiry-based self-efficacy with few engagements in previous professional developments that did not present sufficient opportunities for the instructors to engage in, observe, discuss, and overcome their fears of inquiry pedagogy. This is another significant contribution of our study to teaching practice and literature. In the present circumstances, most of the instructors outside the professional development context will continue to have low self-efficacy of inquiry-based but high self-efficacy of traditional science teaching. This is an impediment to inquiry practices in classrooms, and compromise the implementation of inquiry curricula initiatives.

Findings from this study also show that most instructors outside the professional development context had high PSTE based on traditional understandings of teaching, while others had high PSTE based on partial understandings of inquiry. Instructors with traditional and partial understandings of science teaching equate hands-on activities with inquiry (e.g. DeCoito & Myszkal, 2018; Lardy & Mason, 2011; Mohammed & Amponsah, 2021), are unable to differentiate school science from actual science, and make vague and inaccurate references to the roles of teachers in inquiry pedagogy (Mohammed & Amponsah, 2021). Teachers’ levels of self-efficacy alone are insufficient to produce successful enactments of inquiry-based teaching (Bandura, 1977). Teachers must also have adequate knowledge and skills of inquiry pedagogy to be successful in classroom. In this regard, Lardy and Mason (2011) noted that:

When teachers feel that they are teaching science the way it should be taught, but actually hold images of effective science teaching that are different from those of science education researchers, they engage in ineffective instruction that misaligns with their high self-efficacy. (p. 15).

Clearly, statements from theory and literature suggest that most of the instructors outside the professional development context overestimated their self-efficacy in inquiry teaching, and must have likely been engaged in ineffective teaching that misaligns with their high self-efficacy. In the current situation where the instructors have overestimated their self-efficacy, they are less likely to reflect on their practices and less likely to reform their pedagogy, even if changing their practices will promote students’ science achievements (Lardy & Mason, 2011).

Additionally, findings from this study show that there were statistically substantial school type, school location, and academic qualification differences in science instructors’ self-efficacy outside of the professional development context. Specifically, there were statistically significant differences in personal science teaching efficacy (PSTE) and science teaching outcome expectancy (STOE) of instructors outside the professional development context, based on school-type and academic qualification. Private JHS instructors outside the professional development context had lower PSTE and STOE of inquiry-based but higher PSTE and STOE of traditional teaching practices than their counterparts in public JHSs. Likewise, pre-university certificate instructors outside of the professional development context had lower PSTE and STOE of inquiry-based but higher PSTE and STOE of traditional teaching practices than graduate instructors outside the professional development context. Similarly, urban JHS instructors outside the professional development context had lower PSTE of inquiry-based but higher PSTE of traditional teaching practices than their counterparts in rural JHSs.

These differences arise because contextual factors under which science teaching occurs influence how instructors cognitively process information (Bandura, 1977). Although didactic science teaching, learning, and assessment practices are generally prevalent in many schools, these practices are more pronounced in certain school contexts (Mohammed et al., 2020) and among certain sub-populations of instructors than others. The manner instructors in different school contexts and sub-populations process didactic teaching, learning, and assessment information partly accounts for the significant differences in self-efficacy of instructors outside of the professional development context. Additionally, the types of support for science teaching that instructors receive in schools they are working in and the types of challenges they encounter from students, administrators, and the educational system is also part of the account of significant differences in self-efficacy of instructors outside the professional development context (McKinnon & Lamberts, 2014). This finding is contrary to most findings obtained from participants in the contexts of professional developments and methods courses. Findings obtained from participants in the contexts of professional developments and methods courses showed that teachers’ perceptions of their school climate and collegial support had significant influence in changing their PSTE and STOE (Lotter et al., 2016). This finding is another substantial contribution of this study to the current literature and teaching practice, in that, school-type, school location, and academic qualification differences in PSTE and STOE of instructors outside of the professional development context have not been sufficiently reported in published journals over the past decade.

While social cognitive theory (Bandura, 1977) states that teachers must hold high self-efficacy as prerequisite for the implementation of inquiry-based teaching practices, curriculum implementation theory (Rogan & Grayson, 2003) suggests that it is unrealistic for diverse teachers in various schools to engage in implementation of the same or similar inquiry practices at the same time. Based on differences in personal science teaching efficacy (PSTE) and science teaching outcome expectancy (STOE) of diverse teachers (pre-university certificate, graduate, and postgraduate) in various schools (private, public, rural, and urban) observed in this study, it is important to adopt differentiated and phased approach to implementing inquiry-based science teaching in many schools. Different instructors in different schools with different self-efficacy levels may start implementing inquiry-based practices with different content knowledge, teaching skills, and progress at different rates.

Conclusions

This study explored science teaching self-efficacy outside of the professional development context. It also combined concepts from social cognitive and curriculum implementation theories and findings from the study to explore effective ways of implementing inquiry-based practices in schools. The findings show that instructors outside of the professional development context had low self-efficacy of inquiry-based but high self-efficacy of traditional science teaching practices. The findings also show that while most instructors outside the professional development context had high PSTE based on traditional understandings of teaching, many others had high PSTE based on partial understandings of inquiry. Again, the findings show that there were statistically substantial school type, school location, and academic qualification differences in self-efficacy of science instructors outside the professional development context.

Based on combination of social cognitive theory (Bandura, 1977), curriculum implementation theory (Rogan & Grayson, 2003) and findings from the study, it was ascertained that differentiated and phased approach to implementation of inquiry-based teaching practices needs to be adopted in various schools of diverse teachers. Different teachers in different schools with different levels of self-efficacy may start implementing inquiry practices with different content knowledge, teaching skills, and progress at different rates.

Based on the findings, we recommend that junior high school instructors in many schools be provided with sufficient and appropriate professional developments to enhance their self-efficacy for inquiry-based science teaching. We also recommend that professional development activities should be tailored based on differences in personal science teaching efficacy and science teaching outcome expectancy of diverse instructors in various schools.

More studies employing pretest-posttest methodologies in the context of professional developments are needed to examine the nature and levels of self-efficacy changes in science instructors in many schools. Again, more studies are needed to examine the levels of significance, if any, of school type, school location, and academic qualification differences in PSTE and STOE of science instructors in professional development contexts.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Salifu Maigari Mohammed

Dr. Salifu Maigari Mohammed is a professionally trained and certified science teacher. He has over two decades of experience in teaching Integrated Science, Physics, and Mathematics in classrooms. He also has over a decade and half experience in assessing students’ learning achievements in Integrated Science and Physics, as an external examiner. Dr. Mohammed is interested in research about and application of innovative pedagogies in promoting meaningful learning and scientific literacy among students. He is also interested in research about science education for Second Language Learners (SEL), and influences of affective attributes in science education.

Austin Wontepaga Luguterah

Dr. Austin Wontepaga Luguterah is a trained professional teacher. He holds Teacher’s Certificate ‘A’, BED in Physical Education and English Language, MPhil in sports management, and PhD in Public Administration with specialty in Sports governance and policy. He is interested in research about sports management, governance, policy, and teacher education.