Teachers in science education play a crucial role in modeling science and engineering practices (SEPs) for effective scientific inquiry since the release of the Next Generation Science Standards (NGSS). This research explores how teachers model the process of engaging in instructional practices that draws on sociocultural learning theories by focusing on their demonstrations and exemplifications of essential questioning, problem-solving, and thinking approaches to scientific discovery. Field notes from 363 middle school teacher observations were deductively coded based on the eight SEPs outlined in A Science Framework for K-12 Science Education. Quantitative Ethnography (QE) and Epistemic Network Analysis (ENA) statistical modeling were employed in order to provide a deeper insight into the relationships among SEP codes and visualize their connections as they are displayed by teachers. Results show the most commonly used practice modeled across teachers was Constructing explanations and designing solutions, while the practice for Engaging in argument from evidence was less frequently seen. Differences were seen between math and science teachers in which SEPs most often occurred together, with Using mathematical and computational thinking and Asking questions and defining problems predominately co-occurring in math classes, while science teachers demonstrated the practices of Constructing explanations and designing solutions in tandem with Obtaining, evaluating, and communicating information more frequently. This study underscores the critical need of teachers' familiarity and comfort in using and modeling the SEPs required for scientific inquiry in order to enhance students' scientific and engineering capabilities.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
With the national move toward the adoption of Next Generation Science Standards (NGSS; National Research Council [NRC], 2012), the science classroom relies on the essential role of teachers in modeling the foundational practices necessary for scientific inquiry. One way teachers accomplish this task is by demonstrating the specific skills, methods, and approaches that students need to use in the context of science and engineering activities and investigations, thus providing them with clear examples to follow and helping them to understand how to effectively apply these practices (Abraham, 1998). By observing their teachers' modeling of scientific inquiry (Schwarz, 2009), students gain insights into how to perform these practices and enhance their own scientific and engineering capabilities. Building upon research in sociocultural learning theories, this study investigated the dynamic interactions as educators exemplified the process of engaging in scientific practices and problem-solving to explore the natural world. The overarching aim examines the dynamic interactions in STEM classrooms, with a specific focus on the pivotal role of teachers in modeling and transferring practices from science and engineering. We employed the use of Epistemic Network Analysis (ENA), which leverages statistical analysis and modeling visualizations to provide an enhanced understanding of connections between phenomena (Shaffer, 2017) and is becoming a statistical method of choice in educational research (Elmoazen et al., 2022).
Science and Engineering Practices
NGSS were created in response to a call for a quality science education that aligns with research and deepens student understanding of science-based fields (NRC, 2012). The NGSS provide a framework for implementing high-quality research-based science and engineering principals in the classroom. Built as a three-dimensional perspective of science, NGSS incorporate crosscutting concepts (CCCs), disciplinary core ideas (DCIs), and science and engineering practices (SEPs) to provide students with a roadmap to learning while simultaneously giving teachers a strategy for instruction. Implementation of NGSS and SEPs across K-12 education in the United States has helped shape the educational landscape. Previous research identified engineering practices as missing within the primary and secondary classrooms (Moore et al., 2015), but they have more recently become integrated at the elementary (Kang et al., 2019; Lilly et al., 2022; Merritt et al., 2018) and middle/high school levels (Colclasure et al., 2022; DeLisi et al., 2021). SEPs are integral to student engagement as they work to understand complex phenomena and are specifically aimed to incorporate the field practices implemented by scientists and engineers (Next Generation Science Standards, 2013), and these are especially critical in higher grades. A 2019 report stated that science and engineering investigations “should be the central approach for teaching” in grades 6 through 12 (Brenner et al., 2019), enabling middle and high school students to perceive themselves as engineers and scientists.
Anzeige
A unique component of SEPs is their sequential nature, guiding students from asking questions to communicating findings (Table 1), while also allowing for practices to overlap and interconnect. SEPs integrate real-life scenarios and help students with exploring the natural world, while developing the practices that allow for scientific knowledge construction and the application to real-world problems and challenges. When outlining these necessary components of scientific inquiry, NGSS developers noted that the choice of the word ‘practices’ versus ‘skills’ was intentional (NRC, 2012, pg. 30). Thus, embedded within the SEPs are integral pieces for employing computational thinking skills (Lilly et al., 2022) and inquiry-based practices (Alston et al., 2020). Moreover, the NGSS and SEPs are designed to integrate math into the science curriculum (Marshall et al., 2017; Weintrop et al., 2016), as math plays a critical role in the understanding of relationships between scientific concepts (Furner & Kumar, 2007).
Table 1
Codebook based on NGSS SEPs
Cohen’s kappa
Condensed Code Definition
Example [line]
SEP.1 Asking questions and defining problems
k = 0.88
Science questions lead to descriptions
and explanations of how the world
works, which can then be empirically tested. Engineering questions clarify problems, seek solutions, and identify constraints
Asked the students to think through things they “know” and what they “need to know”, in order to [design and build an assembly line]. … “That is the question you need to think of ‘How do you want to load this – from the top, bottom, or the side?’ And you are designing this to make it move 6 blocks autonomously. What does autonomously mean? (student answered). So, what this looks like is up to you guys.” [456]
SEP.2 Developing and using models
k = 0.79
A practice of both science and engineering is to use models as helpful tools for representing ideas and explanations. These tools include diagrams, drawings, physical replicas, mathematical representations, analogies, and computer simulations
Students were asked to build simple circuits using a Gizmos simulation. Students individually"tested"a variety of materials and classified these materials as"conductors"or"not conductors."Students were tasked with a challenge:"I want you to try this challenge. On your own, create a circuit using three wires, a switch, a battery and a bulb. Is there a conductor that works best?" [811]
SEP.3 Planning and carrying out investigations
k = 0.72
Scientists and engineers plan and carry out investigations in the field or laboratory. Investigations are systematic and require clarifying what counts as data and identifying variables. Engineering investigations identify the effectiveness, efficiency, and durability of designs under different conditions
Since a key focus of the lesson was precision and accuracy in measurement, [teacher] often reminded students to take actual measurements."Guys, when you purchase materials at the store, do you always trust the stickers? Is a 2″ X 4″ actually 2 inches by 4 inches?"He put a wood block on the table. Two students came over with calipers and a ruler; they began to measure the block."Hey, it's only 3 1/2 by 1 1/2.”"That's right,"replied [teacher],"a 2″ X 4″ isn't really 2″ X 4″. Measure everything.” [92]
SEP.4 Analyzing and interpreting data
k = 0.67
Because data patterns and trends are not always obvious, scientists use a range of tools—including tabulation, graphical interpretation, visualization, and statistical analysis—to identify the significant features and patterns in the data. Engineering investigations include analysis of data collected in the tests of designs to compare different solutions
[In a] simulated prey/predator activity. … [The teacher] stated,"You probably are beginning to see some trends in some that lasted longer. Your greens may have lasted. You may not have many green ones, or you may have known to look for them. Sometimes predators learn what their prey do to disguise themselves.” [662]
SEP.5 Using mathematical and computational thinking
k = 0.70
Mathematics and computation are fundamental tools for representing physical variables and their relationships. They are used for a range of tasks, such as constructing simulations; solving equations; and recognizing, expressing, and applying quantitative relationships
During the video about perfect squares that used different colors, the teacher asked, “Why are the colors [in the video] changing?” This helped students understand and visualize square roots. [43]
SEP.6 Constructing explanations and designing solutions
k = 0.65
The end-products of science are explanations, and the end-products of engineering are solutions. The goal of science is the construction of theories while the goal of engineering design is to find solutions to problems. The optimal choice depends on how well the proposed solutions meet criteria and constraints
As the students were trying to code their robots to move, the teacher would propose different solutions that they could choose from and help them think through which one would be the better option. [400]
SEP.7 Engaging in argument from evidence
k = 0.72
Scientists and engineers use argumentation to listen to, compare, and evaluate competing ideas and methods. Scientists and engineers engage in argumentation when investigating a phenomenon, testing a design solution, resolving questions about measurements, building data models, and using evidence to evaluate claims
“Which do you think would be the most profitable option?''… Students,"This is a hard decision."Teacher,"We're going to look at the data and graphs for each option and see which would be the most profitable.” [945]
SEP.8 Obtaining, evaluating, and communicating information
k = 0.66
Scientists and engineers must be able to communicate clearly and persuasively the ideas and methods they generate. Communicating information and ideas can be done in multiple ways: using tables, diagrams, graphs, models, and equations, as well as orally, in writing, and through extended discussions
The teacher teaches analytical and creative thinking where she asks students to evaluate different websites and to create a website based on their knowledge and their own interest. The teacher also provided students [an] opportunity to engage in research-based thinking where they explore and review different models (website created by other students). [609]
ENA
Due to the multi-dimensional design of NGSS, ENA is a valid tool for visualizing the dimensions, relationships, and connections between concepts and can be useful in quantitatively capturing the complexity of instructional practices in STEM education. ENA has been used in science research and NGSS around the use of assessments (Mulvey et al., 2021; Talafian & Kang, 2022), student learning (Bressler et al., 2019; Dabholkar et al., 2020), and the standards themselves (Siebert-Evenstone & Shaffer, 2019). In regard to SEPs, ENA has been recently used to understand teacher ideas around engineering practices (Parrish et al., 2022) and to analyze the presence of SEPs in a science curriculum (Peel & McGee, 2023). However, empirical research using ENA on teachers practices in the classroom is limited, thus opening a novel trajectory of research.
Theoretical Framework
Building on the work of Vygotsky (1978), sociocultural situatedness is imperative to learning, where student learning is guided by the modeled actions and practices of teachers. Teacher practices commonly utilize modeling and scaffolding to equip students with the ability to carry out and complete tasks that they otherwise would not have the ability to do (Wood et al., 1976), and teachers’ frequently employ this strategy during whole-class instruction (Smit et al., 2013). As teachers model the process of critical thinking and problem solving in tandem with scientific practices (Dabholkar et al., 2020), it is ultimately the student acquisition and uptake of these practices that will translate to later success.
The Current Study
The strength of understanding and using SEPs in the middle school classroom enables teachers to embody how students can perceive themselves as engineers and scientists. However, how middle school teachers are actually demonstrating and implementing this in the classroom is not well understood, opening up research opportunities in SEPs in this population. This study sought to answer the following research question: How do teachers model questioning, higher-order thinking, and problem-solving in STEM classrooms to facilitate scientific inquiry through science and engineering practices in the middle school classroom? In answering this question, we aimed to enhance our understanding of the instructional dynamics that shape students'acquisition of essential skills and problem-solving abilities in the realm of STEM education.
Anzeige
Methodology
Participants
This study recruited 184 middle school teachers from 44 schools in 15 districts across a southeastern state.
Over a span of 13 months, a total of 363 observations were conducted. Middle school teachers in this study were largely white females (81.5% white; 16.9% black; 1.6% other; 85.9% female; 14.1% male). Participants were observed teaching science, math, social studies, English language arts, or electives, such as visual arts and computer science. Participating teachers were observed in classes with varying grade dynamics; 323 (89%) of the observations occurred in classes consisting of same-grade students in sixth, seventh, or eighth grade and 40 (11%) classes included multiple grade levels.
Observation Protocol
Classroom observations are commonly used as part of teacher evaluations to better understand and improve teaching practices (Pianta & Hamre, 2009) and observation protocols allow for general agreement of the constructs of high-quality instruction (Praetorius & Charalambous, 2018). Classroom observations were completed using a validated rubric for teaching and learning standards (National Institute for Excellence in Teaching [NIET], 2021), which establishes the expectations for what classroom-based teachers should know and carry out as integral parts of their instructional practice (National Council on Teacher Quality, 2021). The rubric used was the state-approved teaching metric and was therefore well-understood by both teachers and observers; however, outcomes from the observations were only used for research purposes and were not evaluative in any way.
As this study investigated how teachers model scientific inquiry practices, three categories from the rubric were selected for investigation. The categories included Questioning, Thinking, and Problem-Solving, which align with the SEPs developed by NGSS. Observers looked for instances of the following: Questioning, where the teacher used knowledge and comprehension, application and analysis, and creation and evaluation questioning types; Thinking, where the teacher displays analytical, practical, creative, and research-based thinking; and Problem-Solving, where the teacher teaches abstraction, categorization, drawing conclusions, and prediction of outcomes (NIET, 2021). Teachers were randomly assigned a trained observer for a full class period who documented evidence of demonstrated questioning, thinking and problem-solving. Observers were not tasked with noting when these practices occurred; therefore, field notes did not include the sequence or order of when these practices throughout the class period.
Data Segmentation and Qualitative Coding
Field notes from classroom observations were captured via Qualtrics and deductively coded according to the eight SEPs. Code definitions consisted of the NGSS description (Table 1) and were refined using the NGSS practices specific to middle school grades 6–8 (National Science Teaching Association, 2014) to match the observed population of middle school teachers and provide construct validity.
Data from field notes were organized systematically into lines, stanzas, and conversations using Quantitative Ethnography (QE; Shaffer, 2017; Shaffer et al., 2016), which allowed for modeling visualizations and statistical analysis to provide a deeper insight into the relationships among codes. Segmentation of the data prior to coding organized the data following the hierarchical structure of QE, with a conversation forming the largest unit of analysis to a line being the smallest unit of analysis. Observers had originally separated their field notes based on the the categories of the rubric, Questioning, Thinking, and Problem-Solving, and this categorization enabled a natural split, with each rubric category becoming a conversation. Within each conversation there were 363 stanzas, corresponding to the notes from each teacher observation for that category, resulting in a total of 1089 lines.
After segmentation, coding of the data occurred with four of the authors pairing up, each pair coding four of the SEPs. After reviewing the first 15–20 lines in pairs to align codebook interpretations, one-third of the data (381 lines) was randomly allocated to a test set each person coded independently. Next, concurrent validity was determined through inter-rater reliability of the test set (Shaffer and Ruis, 2021), and Cohen’s Kappa was determined (Cohen, 1960). For codes that did not meet the kappa threshold of 0.65 (Shaffer et al., 2016), the pair returned to the data and discussed disagreements within the coding, followed by a repeated test set of 100 lines and an additional kappa calculation. Once all four codes for the pair met the threshold, each individual in the pair was responsible for independently coding the remaining lines for two of the SEPs.
ENA Modeling
The ENA webtool (Marquart et al., n.d.) was used to map codes according to the selected unit of analysis in a two-dimensional space. In ENA visualizations, codes are represented through filled in circles (nodes), with the frequency of code occurrence shown in the size of the node. Connections between nodes, or the frequency of code co-occurrence, are shown through the thickness of the lines (edges), with line thicknesses corresponding to measured edge weight. Line weights demonstrate magnitude and are directly calculated from the codes, corresponding to the relative occurrence for that unit (Arastoopour Irgens et al., 2020); they can be interpreted as a percentage and therefore used as a measure of comparison. These visualizations, or networks, provide a visualization of co-occurrences of codes through weighted lines connecting a pair of codes. ENA presents networks within a two-dimensional space on an x- and y-coordinate plane (D1 and D2, respectively). While the placement of the nodes and the orientation of the quadrants does not demonstrate a positive or negative direction, it allows for relationships between codes to be inferred (Shaffer et al., 2016).
ENA allows for within-group and between-group comparison. In single group models, the center of mass of the mean network for all unit networks is located at the origin, with code nodes placed in a fixed space creating an interpretation of the space itself. For comparison models, ENA calculates a mean network for each group and the centroid of the mean represents a statistic for each group that is then placed in a multi-dimensional space (Shaffer et al., 2016). Group difference networks are modeled by calculating the mean for each group and subtracting the mean networks, and the resulting difference network displays the stronger connection between the two codes of the two groups. Correlation coefficients for both single group and between-group models measure the strength and direction of the relationships between the codes.
For all models used in this analysis, a whole-conversation stanza window was selected as it demonstrated the connections related to categories of teacher practices (Zörgő et al., 2021) as temporality of events was not considered. Two models resulted from the data; model 1 demonstrates the network for all observed teachers and model 2 includes the networks for only science and math subject areas teachers (Table 2). Both Pearson and Spearman correlation coefficients suggest a strong positive relationship between the codes, indicating teachers exhibited similar practices while modeling SEPs.
Table 2
Model and network parameters
Model 1
Model 2
Mean Network
Science-Math Subtracted Network
Unit of Analysis
Teacher
Subject > Teacher
Conversational Unit
Conversation
Conversation
Moving Stanza Window
Whole conversation
Whole conversation
SVD1 Variance
16.4%
16.5%
SVD2 Variance
11.9%
11.5%
Pearson Correlation D1
0.94
0.94
Spearman Correlation D1
0.95
0.95
Pearson Correlation D2
0.89
0.87
Spearman Correlation D2
0.89
0.89
Results
This study aimed to investigate how middle school teachers model questioning, higher-level thinking, and problem-solving skills with science and engineering practices in the STEM classroom.
Model 1—Mean Network for SEPs in Teacher Practices
The first ENA model included middle school teachers across all subject areas for the eight SEP codes. Examining the mean network for SEPs at the individual teacher level, evidence of each SEP connected to the other SEPs was seen (Table 3). While all possible connections between codes were seen, differences in connectivity patterns and frequency of code occurrence are visualized in Fig. 1, with 28.3% of the variance explained by this model. The most prominent code seen across all observations was SEP.6, Constructing explanations and designing solutions, which had equivalent connections between the nodes of SEP.1, SEP.2, and SEP.8. The frequent co-occurrence of SEP.6 with these other three codes indicates a common practice among these middle school teachers in demonstrating how they construct their explanations and solutions (SEP.6), while modeling how they ask questions in order to define their problems (SEP.1), how they use models as a foundation in the process (SEP.2), and how they communicate an understanding of their problem solution (SEP.8). One example of this was seen in the field note comment:
The teacher asked questions consistently throughout the lesson. She varied the questions from knowledge and comprehension (What is frostbite? What is an insulator? Tell me what a conductor is?) to application and analysis (Why do you think a flashlight would be helpful? Did we talk about why that was important? Why was it?) to creation and evaluation (evidenced by student presentation) and also by student-created argument posters displayed in the room. Students generated questions as evidenced by the inquiry presentation (self-directed learning) and also in asking peers about their reasoning for ranking survival objects. [294]
Here, the teacher can be seen modeling all four of these SEPs in tandem during one lesson. First, the teacher began with a variety of guiding questions students should ask (SEP.1), starting with “What?” and building to “Why?”, then scaffolding to questions generated by the students themselves. She used a flashlight as a model for the lessons’ concept (SEP.2) and provided time for students to communicate information through “student presentations”, but also to evaluate ideas through the use of “argument posters” (SEP.8) It was noted that after modeling all three of these SEPs, the teacher then turned those practices back to the students to begin implementing themselves, seen in the statement of “self-directed learning”.
Countering the salient connections of SEP.6, SEP.7, Engaging in argument from evidence, overwhelmingly had the weakest connections to all other nodes, with all seven connections having the smallest line weight seen in the data (≤ 0.06). Low line weights connecting nodes indicate an infrequent occurrence of those codes co-occurring in the data. However, while low, these connections from the data were still seen in teacher practices. One example can be seen between SEP.7 and SEP.4, Analyzing and interpreting data:
Analyzed graphs to answer questions about phone use. T [Teacher]- This person is getting 10 cents on a dollar more. What about the other person? T- Look at the information on the graph. Who'll run out of money first? S [Student]- jan. T- See how this helps you in real life? If you're running a business, you will have to make decisions based on comparisons. T- What kind of equation can you write now? You have intersect and the slope. [616]
In this math class, the teacher modeled practices on how to extract information from graphs (SEP.4) and analyze data, while also helping them construct an argument based on this information (SEP.7) as evidenced by the statement that running a business in real-life requires them to “make decisions based on comparisons.” While this teacher connected and demonstrated these practices, the overall weak connection between these nodes indicates an infrequent co-occurrence of these codes in the data.
Model 2—Science and Math Teachers
Further analysis examined which SEPs were modeled within specific discipline-area classes. In this second model, teachers for subject areas except math (n = 71) and science (n = 88) classes were removed to allow for a subject-specific investigation of science and engineering practices in the STEM classroom. Figures 2, 3 and 4 plot the networks for both math (blue) and science (red) teachers on the same two-dimensional plane. Through two-sample t-tests assuming unequal variance, significant differences were seen in between groups on both the x- and y- dimensions (x: t(137.55), p = 0.00, Cohen’s d = 1.18; y: t(153.29.), p = 0.00, Cohen’s d = 0.67).
Fig. 2
Mean epistemic network for SEPs demonstrated by science teachers (right, red), where the colored square is the mean of each group and the dashed boxes demonstrate the groups' 95% confidence intervals
Collectively, the science teachers made connections between all eight SEPs during their classroom instructions and activities, as shown in the mean science network (Fig. 2). Similar to the pattern seen in Model 1, they most commonly made connections to SEP.6, Constructing explanations and designing solutions, with SEP.6 being more strongly connected to SEP.8, Obtaining, evaluating, and communicating information (0.19), SEP.2, Developing and using models (0.17), SEP.1, Asking questions and defining problems (0.16), SEP.3, Planning and carrying out investigations (0.14), and SEP.4, Analyzing and Interpreting Data (0.11). The highly connected SEP.6 displayed in the network indicates that science teachers readily modeled how to construct explanations and design solutions through their classroom interactions. The teachers most frequently made connections between SEP.6 and SEP.8, as displayed in Fig. 3 by the thick red line connecting these two codes. For example, in the field notes of one teacher, the observer stated:
While students looked at a typical fidget spinner as a model to get an idea of what they needed to do when designing their own, she told them to “feel the weight”, “how does it balance,” and “what causes it to spin” to get them to think about that they needed to incorporate in their own design. She also showed them pictures of some fidget spinners made out of legos to give them an idea. She began with images of very simple lego spinners and told them “This is just to get an idea of what you need. This is spinner 101 and you’re building more than this” but then she showed some other pictures that were more and more advanced to give them both a visual of expectation and idea to start them thinking. [625]
Fig. 3
Mean epistemic network for SEPs demonstrated by math teachers (left, blue), where the colored square is the mean of each group and the dashed boxes demonstrate the groups' 95% confidence intervals
In this statement, the teacher not only provided a model (SEP.2) for the students to look at and obtain information from as they were planning an investigation to build their own spinner (SEP.3), but they also modeled questions a scientist may ask (SEP.1). These questions helped students decide the information they needed to collect (SEP.8) and analyze (SEP.4) before they could begin designing their own fidget spinners (SEP.6). Through this interaction, the teacher displayed the relationships seen among the codes in the data. She provided an example, or model, of how students could start their design process and then guided them through obtaining, evaluating, and communicating information. By modeling how to ask questions, analyze data, and construct explanations of previously designed solutions, the teacher provided appropriate scaffolding to encourage students’ adoption of these practices.
Although not as strong as the connections shown above, science teachers also made connections between SEP.5, and SEP.7, to the other SEPs when engaging their students. These were the weakest connections made by the science teachers in their classroom interaction, indicating that the teachers did not readily model using mathematical and computational thinking using arguments from evidence. However, in one science classroom, it was noted that:
The teacher had valid questions for students that provoked thoughtfulness. She would ask students to take projects home and challenge family members to challenge students'hypotheses. The questioning was pushed beyond the classroom in an applicable way. Students also had questioning for each other and worked towards answering in groups in wanted. [261]
Here, the teacher made connections between SEP.1 and SEP.7 by modeling and asking the students thoughtful questions about their projects, which the students then enacted by asking their peers questions and, although not seen in the classroom, asking the students to engage in arguments from evidence through asking their family members to challenge their current thought processes and hypotheses about their projects.
Network of Math Teachers
Math teachers frequently demonstrated all eight SEPs. Unsurprisingly, code SEP.5, Using mathematical and computational thinking, was the most commonly used code in the network of math teachers, as seen in the thickest lines connecting SEP.5 to the other codes (Fig. 3). The strongest weighted connection in the network of the math classes is between SEP.5 and SEP.1, Asking questions and defining problems (0.26). The co-occurrence of these two codes indicates teachers were exhibiting questioning practices in the classroom as they modeled how to approach a mathematical or computational concept. This connection was evident in the field notes of one math class:
The teacher asked a variety of higher order thinking questions, such as,"how do you know?","what operations are used in the problem,"and"what happens when a one is the denominator of a fraction?"When there is a multiple choice question, the teacher asks,"why is choice B wrong?"As she is going through the steps to solve problems, she is consistently asking the students questions that help them through the problem. [11]
Here, the teacher incorporated a variety of questioning strategies in tandem with mathematical thinking. Questions (SEP.1) centered on information gathering as they identified what math operations they would be completing, seen through the statement, “what operations are used in the problem”. This was then followed by modeling the sequence of computational steps needed in order to solve the problem (SEP.5).
In addition to SEP.1, the network for math teachers also shows frequent connections from SEP.5 to the codes of SEP.2, Developing and using models, and SEP. 6, Constructing explanations and designing solutions. One observer noted evidence for how a teacher displayed all three of these practices in the math classroom:
She kept redirecting students to the tool she provided them, which was the multiplication table. This encouragement to use their resources was a frequent reminder that she gave the class. The class was working on order or operations, which required the students to think about which part of the problem to complete first. When students were stuck on where to start, she started all students at the beginning and walked through what to complete first. Most of her support with students was “What do you do first? Okay, now what do you do? Okay, now what do you do?” step-by-step along the way. [444]
The teacher encouraged the use of a tool, the multiplication table (SEP.2), and to assist with student understanding of the required steps (SEP.5), the teacher modeled questioning practices (SEP.1). This modeling was multi-dimensional, as it included how she decided to start the problem, “What do you do first?”, identified key pieces of information, such as order of operations, and demonstrated sequencing of steps.
Moreover, direct connections between SEP.1, SEP.2 and SEP.6 codes are seen in thick lines found in the left side of the model. Connections between these practices were corroborated in the data itself. On example of this co-occurrence is:
The teacher gave multiple taught math models to students in order for them to decide which one made the most sense to them. The students were allowed to voice their views on the reasoning why they eliminated a person from their project. The teacher fostered these ideas with encouragement. [719]
In this example, the teacher incorporates the practices of SEP.2 and SEP.6. Not only were the students given math models (SEP.2) and encouraged to propose and then explain their solutions using their model (SEP.6), but these were explicitly demonstrated by the teacher.
Subtracted Network of Math and Science Teachers
Figure 4 visualizes the differences between practices of math and science teachers. The teachers’ plotted points show a division at the y-dimension, with math and science teachers dispersed on opposite sides of the x-axis. This division is exemplified in the subtraction network, resulting in an unsymmetric model, with math teachers demonstrating stronger edge weights on the left side of the network; science teachers display the opposite pattern, with stronger edge weights on the right side of the network. The difference network also allows for inferences in teacher practices between disciplines to be made, based on the placement of nodes within the two-dimensional space. SEP.5 is positioned far-left, with SEP.1 and SEP.4 also in the left quadrants, while SEP.8 is found in the far-right with SEP.3. One inference is that math teachers demonstrate how to interact with the data itself, as a common denominator in these left-sided codes are practices for identification (SEP.1) and analyzing relationships and patterns (SEP.4, SEP.5). In contrast, the codes on the right of the model focus more on the process of scientific inquiry, including the conduction (SEP.3) and subsequent communication of investigations (SEP.8), which was more frequently modeled by science teachers. These results demonstrate areas of strength for these teachers, as well as highlight how different STEM discipline areas may be more conducive to demonstrating and modeling scientific practices.
Fig. 4
Network of math (blue) and science (red) teachers with centered means model (left) and subtracted network (right), where the colored dots in the centered means model represent the teachers for each subject area, the colored square is the mean of each group, and the dashed boxes demonstrate the groups' 95% confidence intervals
This study contributes valuable insights to the evolving landscape of STEM education, and underscores the critical role of teachers in transferring the SEPs required in the context of NGSS (NRC, 2012). The connections among all SEPs demonstrate a collective effort put forth by middle school teachers to adopt and implement all eight of these practices in the classroom. Overall, teachers display a strong tendency to help students construct explanations and design solutions across all content areas. Along with strength in frequently demonstrating this practice, teachers displayed a comfortability of using it in tandem with several other practices of science and engineering, with asking questions, defining problems, use of models, as well as the culmination of communicating results and findings. However, there was less of a tendency to model the process of formulating arguments derived from evidence, either in isolation of with other practices. As better integration of science and math in STEM instruction is called for (Furner & Kumar, 2007), here we begin to understand the current state of these practices in these classrooms. Significant differences were seen between science and math teachers, demonstrating discipline-specific modeling practices in STEM. Math teachers readily included practices that required direct engagement with the data itself, while science teachers mirrored the practices of all teachers, frequently pairing practices that allowed for greater emphasis to be placed on the processes in science investigation.
Teachers play a crucial role in transferring these practices for effective scientific inquiry. This work shows that teachers'modeling of SEPs serves as a useful tool in enhancing students'scientific and engineering capabilities, as it fosters evidence-based problem-solving and argumentation and enables students to see themselves in the role of scientist and engineer (Brenner et al., 2019). This modeling of teaching practices demonstrates cohesion in regards to evidence-based end products, conclusions, and solutions being a frequent occurrence in the middle school classroom. These findings align with sociocultural theory (Vygotsky, 1978), emphasizing that teachers’ modeling of SEPs provides situated learning experiences that help students internalize and apply these practices in meaningful ways. The research further suggests a consistent adoption of these practices across middle school grade levels, indicating a promising trend in STEM education.
Implications and Limitations
One challenge with building these skills in students is that teachers must feel competent in explicitly demonstrating these skills. Therefore, this study highlights a need for continued training on NGSS implementation, echoing findings in research with SEPs at the elementary level (Lilly et al., 2022) and teacher use of science inquiry practices (Alston et al., 2020; Dabholkar et al., 2020) in order to further the development of student outcomes in these areas (Marshall et al., 2017). While this work adds to these calls, these results also demonstrate a need for targeted PD, including differentiated support by discipline area and SEP. One limitation of this work is the temporal aspect as the full class time was used as a single event and connections were not made between the timing or sequence of these practices. However, this limitation opens up future research avenues factoring in how teachers temporally tie together SEPs. A second limitation is the unknown length of teaching service, either overall or in these discipline areas, which may impact familiarity and ease in using these practices.
Conclusion
NGSS have become an increasingly important area of research within science education. This study investigates how middle school teachers model the process of engaging in scientific practices to unravel the intricacies of how educators transfer the knowledge and skills involved in these practices to students. As the ultimate goal is students themselves adopting these STEM practices, modeling by teachers is integral to student acquisition, and later use, of these skills. These results add to the growing body of work indicating a needed focus on teacher development within STEM disciplines, in order to develop teachers skilled in these practices. Understanding how teachers are enacting these practices in the classroom helps to better understand where targeted support for teachers is needed, to ultimately impact student’s use of STEM practices themselves.
Declarations
Competing interests
The authors have no relevant financial or non-financial interests and no known conflicts of interest to disclose.
Ethics Approval
Approval was granted by the Office of Research Compliance of Clemson University (IRB2020-396). The procedures used in this study adhere to the tenets of the Declaration of Helsinki.
Consent
Informed consent was obtained from all individuals before their inclusion in the study.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Abraham, M. R. (1998). The learning cycle approach as a strategy for instruction in science. International Handbook of Science Education, (Part 1), 349–362. https://doi.org/10.1007/978-94-011-4940-2_30
Alston, D. M., Marshall, J. C., & Smart, J. B. (2020). Differentiating between the Different levels of inquiry instruction: Classroom dynamics that characterize the quality of inquiry instruction. Science Educator,27(2), 81–91.
Arastoopour Irgens, G., Dabholkar, S., Bain, C., Woods, P., Hall, K., Swanson, H., Horn, M., & Wilensky, U. (2020). Journal of Science Education and Technology,29, 137–161. https://doi.org/10.1007/s10956-020-09811-1CrossRef
Brenner, K., Songer, N., & Moulding, B. (Eds.). (2019). Science and engineering for grades 6–12: Investigation and design at the center. National Academies Press.
Bressler, D. M., Bodzin, A. M., Eagan, B., & Tabatabai, S. (2019). Using epistemic network analysis to examine discourse and scientific practice during a collaborative game. Journal of Science Education & Technology,28(5), 553–566. https://doi.org/10.1007/s10956-019-09786-8CrossRef
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement,20(1), 37–46.CrossRef
Colclasure, B. C., Durham Brooks, T., Helikar, T., King, S. J., & Webb, A. (2022). The effects of a modeling and computational thinking professional development program on STEM educators’ perceptions toward teaching science and engineering practices. Education Sciences,12(8), 570. https://doi.org/10.3390/educsci12080570CrossRef
Dabholkar, S., Arastoopour Irgens, G., Horn, M., & Wilensky, U. (2020, June). Students’ epistemic connectionsbetween science inquiry practices and disciplinary ideas in a computational science unit. International Conference of the Learning Sciences. Retrieved from https://par.nsf.gov/biblio/10199203
DeLisi, J., Kook, J. F., Levy, A. J., Fields, E., & Winfield, L. (2021). An examination of the features of science fairs that support students’ understandings of science and engineering practices. Journal of Research in Science Teaching,58(4), 491–519. https://doi.org/10.1002/tea.21669CrossRef
Elmoazen, R., Saqr, M., Tedre, M., & Hirsto, L. (2022). A systematic literature review of empirical research on epistemic network analysis in education. IEEE Access,10, 17330–17348. https://doi.org/10.1109/ACCESS.2022.3149812CrossRef
Furner, J. M., & Kumar, D. D. (2007). The mathematics and science integration argument: A stand for teachereducation. Eurasia Journal of Mathematics, Science and Technology Education,3(3), 185–189. https://doi.org/10.12973/ejmste/75397CrossRef
Kang, E. J. S., McCarthy, M. J., & Donovan, C. (2019). Elementary teachers’ enactment of the NGSS science and engineering practices. Journal of Science Teacher Education,30(7), 788–814. https://doi.org/10.1080/1046560X.2019.1630794CrossRef
Lilly, S., McAlister, A. M., Fick, S. J., Chiu, J. L., & McElhaney, K. W. (2022). Elementary teachers’ verbal supports of science and engineering practices in an NGSS-aligned science, engineering, and computational thinking unit. Journal of Research in Science Teaching,59(6), 1035–1064. https://doi.org/10.1002/tea.21751CrossRef
Marquart, C. L., Hinojosa, C., Swiecki, Z., & Shaffer, D. W. (n.d). Epistemic Network Analysis, Version 0.1.0. http://app.epistemicnetwork.org, last accessed 2024/02/28.
Marshall, J. C., Smart, J. B., & Alston, D. M. (2017). Inquiry-based instruction: A possible solution to improving student learning of both science concepts and scientific practices. International Journal of Science and Mathematics Education,15(5), 777–796. https://doi.org/10.1007/s10763-016-9718-xCrossRef
Merritt, E. G., Chiu, J., Peters-Burton, E., & Bell, R. (2018). Teachers’ integration of scientific and engineering practices in primary classrooms. Research in Science Education (Australasian Science Education Research Association),48(6), 1321–1337. https://doi.org/10.1007/s11165-016-9604-0CrossRef
Moore, T. J., Tank, K. M., Glancy, A. W., & Kersten, J. A. (2015). NGSS and the landscape of engineering in K-12 state science standards. Journal of Research in Science Teaching,52(3), 296–318. https://doi.org/10.1002/tea.21199CrossRef
Mulvey, B. K., Parrish, J. C., Reid, J. W., Papa, J., & Peters-Burton, E. E. (2021). Making connections: Using individual epistemic network analysis to extend the value of nature of science assessment. Science & Education,30(3), 527–555. https://doi.org/10.1007/s11191-020-00189-5CrossRef
National Council on Teacher Quality (2021, May 11). Expanded ADEPT Support and Evaluation System Guidelines. Retrieved from https://www.nctq.org/dmsView/xxxx. Accessed 28 Feb 2024.
National Research Council. (2012) A framework for K-12 science education practices, crosscutting concepts, and core ideas. National Academies Press, Washington, DC. https://doi.org/10.17226/13165
Parrish, J. C., Pleasants, J., Reid, J. W., Mulvey, B. K., Peters-Burton, E. E., & Recker, A. (2022). Using card sort epistemic network analysis to explore preservice teachers’ ideas about the nature of engineering. Science & Education,33(2), 301–326. https://doi.org/10.1007/s11191-022-00395-3CrossRef
Peel, A., & McGee, S. (2023, October). Trends of Next Generation Science Standards integration in high school science curricula. In Fifth International Conference on Quantitative Ethnography: Conference Proceedings (Vol. 8).
Pianta, R. C., & Hamre, B. K. (2009). Conceptualization, measurement, and improvement of classroom processes: Standardized observation can leverage capacity. Educational Researcher,38(2), 109–119. https://doi.org/10.3102/0013189X09332374CrossRef
Praetorius, A. K., & Charalambous, C. Y. (2018). Classroom observation frameworks for studying instructional quality: Looking back and looking forward. ZDM,50, 535–553. https://doi.org/10.1007/s11858-018-0946-0CrossRef
Schwarz, C. (2009). Developing preservice elementary teachers’ knowledge and practices through modeling-centered scientific inquiry. Science Education (Salem, Mass),93(4), 720–744. https://doi.org/10.1002/sce.20324CrossRef
Shaffer, D. (2017). Quantitative ethnography. Cathcart Press.
Shaffer, D. W., Collier, W., & Ruis, A. R. (2016). A tutorial on epistemic network analysis: Analyzing the structure of connections in cognitive, social, and interaction data. Journal of Learning Analytics,3(3), 9–45. https://doi.org/10.18608/jla.2016.33.3CrossRef
Shaffer, D. W. spsampsps Ruis, A. R. (2021). How we code. In A. R. Ruis, spsampsps S. B. Lee (Eds.), Advances in Quantitative Ethnography (Vol. 1312, pp. 62–77). Springer International Publishing AG. https://doi.org/10.1007/978-3-030-67788-6_5
Siebert-Evenstone, A., spsampsps Shaffer, D. W. (2019). Cause and because: Using epistemic network analysis to model causality in the Next Generation Science Standards. In Advances in Quantitative Ethnography (pp. 223–233). Springer International Publishing. https://doi.org/10.1007/978-3-030-33232-7_19
Smit, J., van Eerde, H. A. A., & Bakker, A. (2013). A conceptualisation of whole-class scaffolding. British Educational Research Journal,39(5), 817–834. https://doi.org/10.1002/berj.3007CrossRef
Talafian, H., spsampsps Kang, H. (2022, October). Modeling students’ performances in physics assessment tasks using epistemic network analysis. In International Conference on Quantitative Ethnography (pp. 285–298). Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-31726-2_20
Vygotsky, L. (1978). Mind in society. Harvard University Press.
Weintrop, D., Beheshti, E., Horn, M., Orton, K., Jona, K., Trouille, L., & Wilensky, U. (2016). Defining computational thinking for mathematics and science classrooms. Journal of Science Education and Technology,25(1), 127–147. https://doi.org/10.1007/s10956-015-9581-5CrossRef
Zörgő, S., Swiecki, Z., spsampsps Ruis, A. R. (2021, February). Exploring the effects of segmentation on semi-structured interview data with epistemic network analysis. In International Conference on Quantitative Ethnography (pp. 78–90). Springer, Cham.
Die im Laufe eines Jahres in der „adhäsion“ veröffentlichten Marktübersichten helfen Anwendern verschiedenster Branchen, sich einen gezielten Überblick über Lieferantenangebote zu verschaffen.
