Science & STEM
Instruction Toolkit

The single biggest idea to understand about modern science teaching is that it has shifted from science as a noun (a body of facts to absorb) to science as a verb (a set of practices to engage in). Instead of a teacher delivering information and students memorizing it, today's approach has students doing the authentic work of scientists and engineers: asking questions, planning and carrying out investigations, analyzing data, constructing and revising models, arguing from evidence, and designing solutions to problems — all in the service of making sense of phenomena (how and why things happen in the world). STEM takes this further by weaving together science, technology, engineering, and math (plus arts in STEAM), because real-world problems are inherently interdisciplinary. This isn't about abandoning content — students still learn deep science ideas — but about learning that content through doing science rather than apart from it (the two, research shows, are inseparable). The result is science that's more engaging, more authentic, and more aligned with how knowledge is actually built. See NSTA.

If there's one shift that captures the transformation of science education, it's the move from memorizing to sensemaking. The old model — lecture, textbook, memorize vocabulary, recall it for the test — treated students as passive recipients of established facts, and treated any hands-on 'inquiry' as an occasional side dish to the 'real' work of absorbing content. The developers of the Framework for K–12 Science Education call the new vision 'perhaps the most significant shift': students should make sense of phenomena or design solutions to problems — figuring out how and why the world works rather than just learning about it. And a crucial, research-based insight underpins this: you cannot learn scientific content separate from engaging in the practices of science — the two are inseparable, so 'sensemaking through doing' isn't a nice add-on but the very mechanism of deep learning. In practice, this means students spend far less time copying notes and far more time investigating, reasoning, and explaining — building understanding they own and can transfer. Once you internalize this shift, everything else in this toolkit follows from it. See Ambitious Science Teaching.

The Next Generation Science Standards (NGSS), built on the 2012 Framework for K–12 Science Education, are organized around three-dimensional learning — the idea that genuine science understanding requires three dimensions working together. The Science and Engineering Practices (SEPs) are what scientists and engineers actually do (§05). The Crosscutting Concepts (CCCs) are big ideas — like cause and effect, systems, and patterns — that cut across every scientific discipline and serve as lenses for making sense of anything (§06). And the Disciplinary Core Ideas (DCIs) are the foundational content of each field (the physics, chemistry, biology, and earth/space science ideas). The critical insight — emphasized by the Framework's own authors — is that these three dimensions must be integrated, not separated: the practices and crosscutting concepts aren't merely 'in service of' teaching the content; rather, all three combine as students work to make sense of phenomena or design solutions, which drives the whole learning process. NGSS expresses this through 'performance expectations' that bundle all three dimensions together. Three-dimensional learning is the foundation of everything modern science teaching aspires to. See NextGenScience.

The Science and Engineering Practices (SEPs) are the heart of what makes modern science classrooms look different: instead of watching science being explained, students engage in the practices scientists and engineers actually use. There are eight: (1) asking questions and defining problems; (2) developing and using models; (3) planning and carrying out investigations; (4) analyzing and interpreting data; (5) using mathematics and computational thinking; (6) constructing explanations (for science) and designing solutions (for engineering); (7) engaging in argument from evidence; and (8) obtaining, evaluating, and communicating information. Notice that these are verbs — active work students do, not information they receive — and that they span the full arc of doing science, from wondering about something to communicating conclusions. The practices aren't a separate 'inquiry unit'; they're the means through which students learn content (§03). When you plan a lesson, a powerful question is: which practice(s) will my students actually do today? Getting students engaging in these practices regularly is the single biggest lever for authentic science learning.

The Crosscutting Concepts (CCCs) are the least familiar of the three dimensions — teachers often call them 'the most feared' — but they're genuinely powerful once understood. They are big ideas that cut across every scientific discipline, and there are seven: patterns; cause and effect; scale, proportion, and quantity; systems and system models; energy and matter; structure and function; and stability and change. What makes them valuable is that they function as mental tools or lenses for sensemaking — ways of looking at any phenomenon that help students ask productive questions and see connections. A student examining an ecosystem, a chemical reaction, and a machine can use the same lens of 'systems and system models' to make sense of all three, revealing that science isn't a collection of disconnected topics but a web of recurring patterns of thought. The teaching move is to make these lenses explicit and repeated: prompt students to notice cause and effect, look for patterns, or think about a system, again and again across units, until these become automatic ways of thinking. Far from an abstract add-on, the crosscutting concepts are what let students transfer scientific thinking from one context to the next.

The practical engine of modern science teaching is phenomena-based instruction: rather than opening a unit by announcing the topic ('Today we're starting photosynthesis') and marching through vocabulary, you begin with a phenomenon — a real, observable event or puzzle that sparks curiosity and demands explanation (Why does a plant grow toward light? Why do the northern lights appear?). Students then spend the unit working to explain how and why that phenomenon occurs, using the science and engineering practices, crosscutting concepts, and disciplinary core ideas as the tools to figure it out. A well-chosen anchoring phenomenon can drive an entire unit — students keep returning to it, building and revising their explanation as their understanding grows. This flips the logic of a lesson: content is no longer the starting point to be delivered, but the payoff students construct while making sense of something genuinely interesting. Choose phenomena that are relevant, engaging, and puzzling (ideally connected to students' lives and communities), and the perennial question 'why do we have to learn this?' largely answers itself — because students want to solve the puzzle in front of them.

At the center of doing science is inquiry: students investigating questions the way scientists do — asking questions, planning investigations, gathering and analyzing data, and drawing evidence-based conclusions. Real inquiry is both hands-on and minds-on — the goal isn't just physical activity but active thinking (a student can follow a lab procedure while their brain is switched off, which isn't inquiry). The key craft is finding the right balance of structure and freedom. At one extreme, pure unguided 'discovery' (turn kids loose with materials and hope they discover the concept) tends to be chaotic and ineffective — novices need support. At the other extreme, recipe-following labs where students just execute steps to a known result build little real understanding. The sweet spot is guided inquiry: investigations scaffolded enough to be productive and safe, but open enough that students make genuine decisions — choosing variables, designing an approach, interpreting messy data, grappling with the unexpected. In guided inquiry, students do real scientific thinking within a supportive structure. As they gain skill, you can gradually release more responsibility. The aim throughout: students investigating and reasoning, not just performing steps. See our Project-Based Learning toolkit.

Science is fundamentally something you do, and hands-on labs and investigations let students experience it directly — manipulating real materials, observing real results, and confronting the messiness of the actual world. This matters because concrete experience makes abstract concepts tangible, memorable, and engaging in a way that reading about them never can (a student who has actually watched a chemical reaction understands it differently than one who only read the equation). But hands-on activities carry two important caveats. First, hands-on must be paired with minds-on: the physical activity has to serve sensemaking, not just keep hands busy. Beware 'activity for activity's sake' — a fun lab that students enjoy but from which they extract no scientific understanding is a missed opportunity, so always connect the doing to the ideas (and debrief afterward: what did we figure out, and how do we know?). Second — and non-negotiable — safety comes first: know and follow lab safety protocols, provide proper supervision and safety training, and never compromise on it. Done thoughtfully and safely, hands-on science is where abstract concepts become real, and where many students first fall in love with science. See the Exploratorium for hands-on ideas.

Underlying every technique in this toolkit is one principle: sensemaking — the idea that students should construct scientific understanding from evidence and reasoning, rather than simply receive it pre-packaged. This requires a genuine shift in the teacher's role, from teller to facilitator. The instinct to explain — to just tell students the answer — is strong (it's faster, and it feels like teaching), but telling produces shallow, quickly-forgotten knowledge, and it robs students of the intellectual work that builds deep, transferable understanding. Instead, your job is to orchestrate sensemaking: pose questions that prompt reasoning, surface and probe students' thinking, structure discussions where students build on and critique each other's ideas, and use evidence and models to help the class construct explanations together. This is harder and slower than lecturing, and it requires tolerating productive struggle and uncertainty — but it's where real learning happens. The mental discipline for the teacher is to keep asking 'how can I get students to figure this out?' rather than 'how can I explain this?' When students do the thinking, they own the understanding. Sensemaking is the goal that all the practices, phenomena, and inquiry ultimately serve.

A defining practice of science is arguing from evidence — making claims and defending them with data, and critiquing others' claims — and students should learn to do this too. The most useful framework for teaching it is Claim-Evidence-Reasoning (CER): a claim is a statement that answers the question or explains the phenomenon; evidence is the data (observations, measurements, results) that support the claim; and reasoning is the explanation of why that evidence supports that claim (often connecting to a scientific principle). CER gives students a clear structure for constructing evidence-based explanations and moves them beyond 'because the book says so' or 'because I think so.' Two important teaching points. First, embed CER in genuine scientific discourse — get students talking: discussing, questioning, critiquing, and building on each other's claims respectfully, because argumentation is fundamentally social and talk is where scientific reasoning develops. Second — and often missed — focus on the process, not on students being 'correct.' The goal is to support students in backing up their claims with evidence and explaining their reasoning, and their sophistication will grow with age and experience; a well-reasoned claim from solid evidence is valuable even if the conclusion isn't perfect. Argumentation turns students from answer-givers into evidence-based thinkers — a skill far beyond the science classroom (see our Media & Information Literacy toolkit).

One of the most powerful science practices — and one of the eight SEPs — is developing and using models. In science, a model is a representation (a diagram, a physical build, a drawing, a simulation, a mathematical equation) that helps explain, represent, and predict how a phenomenon works. Modeling is central to how scientists actually reason — they build mental and physical models of invisible or complex systems — and it's transformative for students because constructing their own model forces them to make their thinking explicit and coherent. A key distinction: having students build their own models is far more powerful than handing them a polished textbook diagram to memorize — the intellectual work of creating and refining a model is where the learning lives. Even better is treating modeling as an iterative process across a unit: students draw an initial model of a phenomenon early (revealing their current thinking, including misconceptions), then revise it repeatedly as they gather evidence and deepen understanding — so an evolving model becomes a visible record of growing understanding. Modeling also gives you a window into student thinking (initial models expose what students actually believe). Use models throughout, not as a one-off — they're among the richest tools in science teaching.

A major addition in modern science standards is engineering — the 'E' in STEM — now woven into K–12 science. Engineering is a natural complement to science: where science seeks to explain the world (understand how and why things work), engineering seeks to solve problems (design solutions to human needs and challenges) — two sides of the same investigative coin. Students engage in engineering through the engineering design process: define a problem (including its criteria and constraints), develop and design possible solutions (brainstorm, plan, build a prototype), and test, evaluate, and optimize — iterating to improve. The most important thing to teach about this process is that it's iterative and that failure is expected: the first design rarely works, and 'failing' — then analyzing why and improving — is engineering (this reframes failure from something to avoid into a productive, normal part of the work, a valuable mindset far beyond STEM). Engineering challenges are typically creative, hands-on, and deeply engaging — students building, testing, and refining solutions to real problems — and they let students apply science concepts in authentic ways. They also connect vividly to real STEM careers (§19). Whether it's designing a bridge, a water filter, or a coding solution, engineering brings science to life through building.

Integrated STEM reflects a simple truth: real-world problems and real scientific work don't respect the boundaries between subjects. STEM integrates Science, Technology, Engineering, and Mathematics (and STEAM adds the Arts), and the point is to connect these disciplines rather than teach them in isolated silos — because a scientist analyzing data uses math, an engineer designing a solution applies science and technology, and understanding a modern problem (climate, health, energy) requires all of them together. As the vision of 'Science for All Americans' put it decades ago, science, mathematics, and technology are interdependent human enterprises. In practice, integrated STEM often looks like project- and problem-based learning (see our PBL toolkit): students tackle a real, meaningful challenge that naturally draws on multiple disciplines. A growing emphasis within STEM is the 'T' — technology, computer science, and coding — which is increasingly essential across every field. Integration makes learning more relevant, authentic, and coherent (students see how it all connects and why it matters), though it takes thoughtful design to keep each discipline rigorous rather than superficial. Done well, integrated STEM mirrors how knowledge is actually built and used in the world.

One of the most important — and humbling — truths in science teaching is that students are never blank slates. They arrive with prior ideas about how the world works, built from years of everyday experience — and many of these are misconceptions: intuitive 'naive theories' that feel obviously true but are scientifically wrong (that heavier objects fall faster, that the seasons are caused by Earth's distance from the sun, that a coat 'makes heat'). The critical, research-backed insight is that simply telling students the correct answer usually does not work — misconceptions are remarkably sticky, and a student will often nod along to the 'right' answer for the test while their original intuition remains fully intact (the two ideas coexisting). Dislodging a misconception requires conceptual change, which means: first, elicit students' prior ideas (find out what they actually think before teaching — you can't address what you don't know is there), and then confront those ideas with evidence and experience that the students themselves grapple with, so they come to see the limits of their existing idea and have reason to revise it. This is exactly why inquiry and the 5E model (which have students explore and question their own ideas, §08) are so powerful — they work with the reality of misconceptions rather than ignoring it. Great science teaching starts by taking students' existing thinking seriously.

A commitment that must run through all of science teaching: science and STEM are for every student. Historically, STEM fields have dramatically underrepresented women, students of color, low-income students, English learners, and students with disabilities — not because these students can't do science, but because of unequal access, opportunity, representation, and a persistent, damaging cultural myth that 'science is for a certain kind of (usually white, male, 'gifted') person.' Dismantling that myth is core equity work: every student is capable of doing science, and the sense-making, evidence-based approach of modern science teaching (which values reasoning over rote recall of facts) actually widens the door. Concretely, pursue equity by ensuring access (labs, materials, technology, and advanced STEM courses reaching all students, not just some schools or tracks), making science relevant (connecting it to students' lives, cultures, and communities so it feels like theirs), fostering representation (helping every student see people like themselves as scientists and see themselves as capable of it), holding high expectations for all, and teaching in culturally responsive ways. Closing the STEM opportunity gap isn't just fair — it's essential for a scientifically literate society and a diverse STEM workforce. Science for all means all. See our Culturally Responsive Teaching toolkit.

A real and widespread problem worth naming: in many elementary classrooms, science gets squeezed out — crowded aside by the intense focus on reading and math, limited by tight schedules, scarce professional development, and many elementary teachers' own lack of confidence or comfort with science. The consequence is that some children get very little science before middle school, which is a genuine loss — because early science matters enormously. Young children are, by nature, natural scientists: endlessly curious, eager to explore, full of 'why' questions — precisely the dispositions science education wants to cultivate — and early, positive science experiences build a foundation of curiosity, confidence, and interest that shapes whether students see science as 'for them' later on. The encouraging news is that great elementary science is very doable. Integrate science with literacy and math (read and write about science, use math to analyze findings — this reinforces all subjects rather than competing with them), lean into hands-on exploration and phenomena (which suit young children beautifully), and — crucially — remember that you don't need to be a science expert: you can wonder and investigate alongside your students, modeling curiosity rather than authority. Protect time for science, nurture that natural wonder, and don't let it get squeezed out.

The ultimate purpose of science education reaches far beyond producing scientists: while only some students will pursue STEM careers, all of them will be citizens, consumers, voters, and decision-makers in a world saturated with science and technology — so scientific literacy is a goal for everyone. A scientifically literate person can understand scientific information well enough to make informed decisions (about their health, their environment, new technologies), evaluate claims and evidence critically (distinguishing sound science from misinformation — a skill deeply connected to media literacy), and grasp the nature of science itself — that it's evidence-based, self-correcting, and a human enterprise, always open to revision as new evidence emerges (understanding how science works is as important as knowing scientific facts). This is why the sense-making, evidence-based approach matters for all students. At the same time, don't neglect relevance and careers: connect science to students' real lives and the wider world, and open their eyes to the vast, growing, well-paid, and diverse array of STEM careers (which many students never realize are options for them). When students see that science both matters for their lives and opens doors to their futures, the question 'why do we have to learn this?' finally has a compelling answer. See our Media & Information Literacy and College & Career Readiness toolkits.

For guidance, NSTA is the professional home for science teachers, NextGenScience hosts the standards, Ambitious Science Teaching focuses on sensemaking, and PhET and the Exploratorium offer simulations and hands-on ideas; Edutopia adds practical strategies. Keep the through-line in view: students should do science — make sense of phenomena, investigate, argue from evidence, model, and design — integrating the three dimensions, and science is for all. Find STEM programs and camps in our programs directory, and pair this with our Math Instruction, Project-Based Learning, and Experiential Learning toolkits. Start at K12academics.com.

Editable versions of these guides are available on request — see §26, Stay Connected.