teaching science class

Instructional Strategies for Science

ByMichael Fallert

Science has always been one of my favorite subjects.

When I was a kid, I didn’t just pay attention during science class. I went looking for more of it. I read encyclopedias and whatever science books I could get my hands on because I wanted to know how things worked. Why did this happen? What caused that? What would happen if you changed something?

I don’t think that kind of curiosity is unusual in children. In fact, one of the strange things about teaching science is that we often begin with students who are already interested in the subject.

Then, somewhere along the way, we sometimes manage to make it boring.

That seems like quite an accomplishment.

I taught science at both the elementary and middle-school levels, and I still think it is one of the subjects that gives teachers the greatest opportunity to combine serious learning with lessons students actually enjoy. You can question things, observe things, build things, take things apart, collect data, argue over explanations, make predictions, test ideas, and sometimes make a perfectly respectable mess while doing it.

But there is an important distinction here.

A fun science activity is not necessarily a good science lesson.

Students can spend forty minutes launching something across a room, mixing colorful liquids, building a tower, or making something fizz and leave class remembering only that the thing fizzed.

The activity is only the beginning.

Good science instruction helps students connect what they did with what they are supposed to understand.

What Good Science Instruction Is Trying to Do

Science is sometimes taught as if there are two choices.

We can teach students a lot of scientific facts, vocabulary, formulas, and definitions.

Or we can let students “do science.”

I don’t think that is a particularly useful choice.

Students need scientific knowledge. There are things they simply need to know. At the same time, science is more than a collection of facts somebody else discovered.

Modern science education emphasizes three things working together: important scientific ideas, concepts that connect different areas of science, and the actual practices scientists and engineers use to investigate questions and solve problems.

Those practices include asking questions, developing models, planning investigations, analyzing data, using mathematics, constructing explanations, arguing from evidence, and communicating information.

That makes sense to me.

We want students to know some science, but we also want them to understand something about how we know what we know.

Start With Something Worth Figuring Out

One of the simplest ways to improve a science lesson is to begin with a question, observation, problem, or phenomenon that students genuinely have a reason to explain.

Why does condensation form on the outside of a cold glass?

Why can you sometimes see the Moon during the daytime?

Why does one object sink while another floats?

Why does salt make ice melt?

Why does a metal spoon feel colder than a wooden spoon when both have been sitting in the same room?

Why do leaves seem to disappear from the ground year after year?

You don’t need an exploding volcano every Tuesday.

Sometimes the best science question is something students have seen a thousand times without ever stopping to think about.

Give them a chance to notice.

Ask them what they think is happening.

Ask them what evidence they would need to figure it out.

Then teach from there.

This is one place where questioning techniques become especially important in science. A good question can expose what students already understand, uncover misconceptions, create curiosity, or push students beyond an answer they memorized five minutes earlier.

“What is evaporation?” may tell you whether a student remembers a definition.

“Where did the puddle go?” can lead somewhere much more interesting.

Hands-On Science Should Also Be Minds-On Science

I have always been a believer in hands-on and experiential learning, particularly in science.

Let students get their hands dirty.

Let them measure something. Sort something. Grow something. Build something. Observe something. Take something apart when appropriate. Put something back together. Collect some actual data.

Science becomes awfully difficult to love when every lesson starts and ends with reading a textbook.

But hands-on learning needs a purpose.

The National Academies has made this point in different ways over the years: simply doing an activity does not guarantee that students understand the science behind it. Students often need teacher guidance to connect an investigation to the concept they are supposed to learn.

That means the teacher still has a very important job.

Before an activity, ask students to predict.

During it, have them observe carefully and record evidence.

Afterward, make them explain what happened.

Then connect those observations explicitly to the scientific idea.

Suppose students build paper airplanes.

They can certainly have fun making and throwing airplanes.

But the science lesson comes from comparing designs, controlling variables, measuring distance or flight time, collecting results, looking for patterns, and trying to explain why one design performed differently from another.

The airplane is not the lesson.

The thinking about the airplane is the lesson.

Teach Scientific Practices Instead of One Rigid “Scientific Method”

When I originally wrote this page years ago, I said teachers shouldn’t skim over “the scientific method.”

I would phrase that differently now.

Students absolutely should learn how scientific investigations work. But science does not always follow one neat sequence of:

Question. Hypothesis. Experiment. Results. Conclusion.

Real investigations can be much messier than that.

A scientist might begin by noticing a pattern in existing data. Another may build a model. Someone else may compare competing explanations, conduct an observational study, analyze samples, simulate a system, or design a series of experiments that creates ten new questions for every one it answers.

Current science standards therefore emphasize broader science and engineering practices rather than treating one step-by-step method as the universal way science happens.

Students should get repeated opportunities to:

  • ask questions and define problems;
  • develop and use models;
  • plan and conduct investigations;
  • analyze and interpret data;
  • use mathematical and computational thinking;
  • construct explanations and design solutions;
  • argue from evidence; and
  • obtain, evaluate, and communicate information.

You don’t have to squeeze all eight into every lesson.

In fact, please don’t.

Pick the practices that make sense for what students are learning.

Find Out What Students Think Before You Tell Them the Answer

This may be one of my favorite parts of teaching science.

Students arrive in your classroom with explanations for how the world works.

Some are correct.

Some are partly correct.

Some are spectacularly wrong.

And some of the wrong ones make perfect sense when you understand what the student has observed.

That is why I like asking students to explain something before teaching it.

Why do we have seasons?

What makes something heavy?

Where does a plant get most of its mass?

Why can we see?

What happens to food after our bodies use it?

Where does the water in a puddle go?

Why aren’t we buried under thousands of years of dead leaves?

You might be surprised by the answers.

Years ago, one of my better eighth-grade students insisted that Earth travels around the Sun once every day.

She wasn’t guessing.

She was absolutely sure.

Once we started talking about it, I discovered that she wasn’t the only student in the room who thought that.

That was useful information.

If I had simply presented a diagram showing Earth’s orbit and moved along, several students probably could have copied the diagram, answered some questions about it, and carried their original misconception right out of the classroom.

Research on science learning has repeatedly found that prior ideas and misconceptions matter. Students don’t necessarily replace an existing explanation merely because a teacher presents the correct one. Good instruction gives them opportunities to reveal, examine, test, and revise those ideas.

So ask first.

Then listen.

Sometimes the wrong answer gives you a better lesson than the right one would have.

Use Models to Make Invisible Science Visible

A great deal of science involves things students cannot directly see.

Atoms.

Forces.

Electric fields.

Food webs.

The inside of Earth.

Molecules moving in a gas.

Energy moving through a system.

Millions of years of geological change.

Models help students reason about these things.

A model can be a diagram, physical object, drawing, mathematical representation, simulation, map, graph, flowchart, or something students construct themselves.

But I think there is an important difference between showing students a model and having students use a model to think.

Instead of always giving students the finished water-cycle diagram, ask them to draw what they think happens to water after a puddle disappears.

Instead of only showing the food web, give students organisms and have them construct one.

Instead of presenting a perfect diagram of the solar system immediately, ask students to model what could produce day and night.

Then compare.

Revise.

Argue.

Improve the model as students obtain more evidence.

The What Works Clearinghouse also recommends connecting concrete experiences with abstract representations and combining graphics with verbal explanations for subjects such as science.

The goal isn’t just to make the page prettier.

The representation needs to help students understand the idea.

Make Students Do Something With the Evidence

Students sometimes believe the purpose of an experiment is to get “the right answer.”

That can produce some interesting laboratory reports.

If the result doesn’t match what they expected, suddenly that ruler reads 14.2 centimeters instead of the 11.7 centimeters that was actually measured.

Science class should help students become more comfortable with evidence that doesn’t cooperate.

Ask:

What does your data show?

How confident are you?

Was there enough evidence?

What pattern do you see?

What else could explain this?

Did every group get the same result?

Why might the results differ?

What would you test next?

Science becomes much more interesting when students realize that data is not merely something you write into a table because the teacher made a table.

The data is supposed to tell you something.

Teach Students to Explain Their Reasoning

I like asking students, “Why?”

I probably asked it enough over the years to be annoying.

But science needs explanation.

A correct answer doesn’t always tell you whether the student understands anything.

Suppose a student tells you that heavier objects do not necessarily fall faster than lighter ones.

Wonderful.

Why?

What evidence supports that?

What would you predict if we changed the objects?

Can you draw or model what is happening?

Can you explain it to someone who disagrees?

The National Academies identifies classroom discussion and explanation as important parts of learning science, and the What Works Clearinghouse has found strong evidence for asking students deep explanatory questions as a general learning strategy.

That doesn’t mean every answer needs to become an essay.

Sometimes one good follow-up question is enough.

Let Students Disagree—But Make Them Use Evidence

Science class should be one of the safest places in school to say:

“I don’t think that explanation works.”

That is not the same thing as:

“I don’t like your answer.”

Students need practice distinguishing between disagreeing with a person and challenging an idea.

If two groups reach different conclusions, don’t automatically rush in and settle the argument.

Ask each group to show its evidence.

What did you observe?

How did you measure it?

Were the procedures the same?

Could one variable explain the difference?

What additional evidence would help?

This is where collaborative learning and critical thinking fit naturally into science.

A productive science discussion isn’t a room full of students trying to guess which answer the teacher wants.

It is students learning that explanations should survive contact with evidence.

Don’t Turn Science Into Vocabulary Class

Science has vocabulary.

A lot of it.

Students do need to learn words such as photosynthesis, erosion, velocity, nucleus, ecosystem, conductor, and equilibrium.

But I’ve seen science instruction become a cycle of:

Copy the definition.

Memorize the definition.

Take the vocabulary quiz.

Forget the definition.

A word becomes much easier to understand when the student already has something to attach it to.

Let students observe water collecting on a cold container before giving them a formal explanation of condensation.

Let them compare materials before introducing conductivity.

Let them watch organisms interact before asking them to use ecosystem vocabulary.

Sometimes vocabulary needs to be taught before an investigation, particularly when students need the words to understand directions or communicate safely.

Other times, the experience can come first.

There isn’t one rule.

The question is whether the vocabulary is helping students think about the science or replacing the science.

Reading and Writing Belong in Science

One reason I have never completely bought the argument that elementary schools “don’t have time” for science is that science provides such a natural reason to read, write, speak, calculate, measure, and interpret information.

Students can read a short text because they need information to solve a problem.

They can write an explanation of a result.

They can compare two sources.

They can create a labeled diagram.

They can make an evidence-based argument.

They can read a graph.

They can write instructions for a procedure.

They can summarize what they learned from an investigation.

Science doesn’t need to compete with literacy every minute of the day.

Sometimes science is the reason for the literacy.

Recent National Academies guidance for elementary science specifically emphasizes investigations and engineering problems as meaningful contexts for learning while still recognizing an important role for text and other sources of information.

Direct Instruction Still Belongs in Science Class

There is a tendency in education to turn useful ideas into camps.

Inquiry or direct instruction.

Teacher-led or student-led.

Hands-on or textbook.

Explicit teaching or discovery.

I don’t think science teachers need to join one of these teams.

There are times when the clearest and fastest way to help students learn something is simply to explain it well.

Model a process.

Demonstrate how to read the graph.

Teach the vocabulary.

Show the equation.

Explain a difficult concept.

Correct a dangerous misunderstanding before anyone lights anything on fire.

Then give students opportunities to apply that knowledge.

The research literature on science instruction does not support treating “inquiry” and “direct instruction” as a simple either-or contest. The useful question is what students are supposed to learn and what kind of support will help them learn it. Scientific investigations themselves often require substantial teacher guidance.

Good science teaching changes gears.

Sometimes the teacher explains.

Sometimes the teacher demonstrates.

Sometimes students investigate.

Sometimes everybody stops and argues about what the evidence means.

Use Demonstrations When a Full Investigation Isn’t Practical

Not every classroom has a laboratory.

Not every teacher has ninety-minute blocks.

Not every school has cabinets full of equipment.

And sometimes letting thirty students independently investigate something would be expensive, unsafe, or simply ridiculous.

Demonstrations still have a place.

But don’t let students become an audience.

Before the demonstration, make them predict.

Ask them to explain their prediction.

Have them sketch what they expect.

Stop midway and ask what they notice.

Show the result and ask them to reconcile it with their prediction.

A five-minute demonstration can produce quite a bit of thinking if students have something intellectually at stake before you reveal what happens.

Use Formative Assessment Throughout the Lesson

Science gives teachers dozens of opportunities to find out what students understand without waiting for a chapter test.

Ask for a prediction.

Have students draw a model.

Give them a phenomenon and ask for an explanation.

Ask them to choose between two competing claims.

Have them interpret one graph.

Use an exit question.

Ask them what evidence changed their minds.

Have them identify what they still can’t explain.

Ongoing assessment is especially useful because science misunderstandings can hide remarkably well behind memorized vocabulary. The National Academies describes ongoing assessment as an integral part of effective science instruction.

If students can define revolution but still think Earth goes around the Sun every day, the definition quiz didn’t tell you enough.

Help Different Students Reach the Same Important Science

Science classrooms contain students with very different background knowledge, reading abilities, language proficiency, experiences, and confidence.

That does not mean we need twenty-seven different science lessons.

It means we may need different ways for students to gain access to the same important thinking.

A student might need:

  • a diagram along with verbal directions;
  • a partially completed data table;
  • sentence starters for an evidence-based explanation;
  • vocabulary previewed before reading;
  • a physical model;
  • additional examples;
  • a smaller set of data to analyze first;
  • strategic peer support;
  • more time to process a question;
  • or a more challenging extension once the central idea is understood.

That is where differentiated instruction should help.

For multilingual learners in particular, science can actually provide some wonderful entry points because objects, demonstrations, diagrams, models, and shared experiences give students something concrete to talk and write about. The science should remain intellectually worthwhile while the language support makes participation more accessible.

Connect Science to the World Students Already Live In

You could probably teach science for an entire year just by looking carefully at an ordinary house.

Why does the refrigerator get warm on the back?

Why does a microwave heat food?

Why does soap remove grease?

Why does a metal pan heat differently from its handle?

Why does bread rise?

Why do batteries eventually stop working?

Why does a phone know which way you’re holding it?

Why does salt damage some surfaces?

Why does condensation collect on windows?

Science is everywhere, but students don’t automatically make those connections.

Point them out.

Better yet, ask students to start finding examples themselves.

Once students begin seeing science outside science class, the subject becomes less like a collection of school topics and more like a way of looking at the world.

Science Doesn’t Have to Be Elaborate

Teachers are busy.

Elementary teachers especially may be looking at their schedule and wondering exactly where this beautiful hour-long investigation is supposed to fit between everything else they have been told they absolutely must accomplish before lunch.

I understand.

Some very good science instruction is small.

Put an ice cube in two different locations and make predictions.

Grow two plants under different conditions.

Drop two objects.

Observe shadows at several times during the day.

Compare how quickly different materials warm.

Measure something for five minutes a day for a week.

Show one photograph and ask students what they notice and wonder.

Give students two possible explanations and ask which is better supported.

Good science doesn’t require a purchase order every time.

It requires students to have something worth thinking about.

A Few Practical Science Strategies by Grade Level

Elementary School

Take advantage of curiosity while it is still practically pouring out of them.

Use observation, simple measurement, classification, models, short investigations, read-alouds and informational text connected to science, outdoor observations, and questions about ordinary phenomena.

Young children are capable of much more scientific reasoning than we sometimes give them credit for. National Academies guidance emphasizes that even young children can participate meaningfully in science and engineering investigations when instruction is appropriately supported.

Don’t wait until middle school to let science become science.

Middle School

This is a great age for pushing students from “I saw this happen” toward “Here is my evidence and here is why I think it happened.”

Increase the sophistication of data analysis, modeling, controlled investigations, explanations, and scientific argument.

It is also a wonderful age for misconceptions.

Trust me.

Students know enough to have elaborate explanations for things and are still often willing to argue enthusiastically about them.

Use that.

High School

Students need increasing independence, but independence does not mean removing guidance.

Give them opportunities to analyze more complex data, evaluate sources, work with mathematical models, design investigations, critique explanations, and connect scientific ideas across units.

At the same time, explicit teaching remains important when students are dealing with difficult concepts and representations.

“Figure it out yourself” is not an instructional strategy.

Common Mistakes in Science Instruction

I think several problems show up repeatedly.

Activity Without Learning

The class did something memorable, but nobody connected it to the scientific idea.

Ask yourself afterward: What should students now understand that they did not understand before?

Too Much Vocabulary Too Soon

Students spend so much effort decoding terminology that the actual phenomenon disappears.

Teach the language they need, but keep returning to the idea the language represents.

Answering Our Own Questions

Teachers ask a really good question.

Nobody answers for three seconds.

We panic.

Then we answer it ourselves.

Give them time.

Treating Every Wrong Answer as Something to Eliminate Immediately

Sometimes a misconception is instructional gold.

Explore it before correcting it.

Calling Something “Inquiry” and Removing All Guidance

Students need support learning how to investigate, analyze evidence, build models, and explain results.

Guidance does not ruin inquiry.

It often makes productive inquiry possible.

Confusing Engagement With Learning

I absolutely want students engaged.

But “they loved it” and “they learned it” are not identical statements.

The goal is both.

If You Want to Improve Your Science Teaching Tomorrow

Don’t try to overhaul everything.

Take one science lesson you already teach.

Find one place where you normally tell students something.

Before telling them, give them a reason to think about it.

Show them something.

Ask for a prediction.

Present a problem.

Ask what they already believe.

Have them draw a model.

Give them two explanations and ask which one they think is better.

Then teach.

At the end, come back to that original idea and ask what changed.

That little cycle—think, investigate or learn, explain, revise—can turn a fairly ordinary science lesson into something much more meaningful.

And perhaps more importantly, it keeps alive the thing that made many of us interested in science in the first place.

The feeling that there is something out there we don’t understand yet.

And that maybe we can figure it out.

Sources and Further Reading

Michael Fallert, retired teacher and founder of InstructionalStrategies.org

Michael is a retired teacher with 24 years of classroom teaching experience and the editor of InstructionalStrategies.org. He focuses on practical, evidence-aware guidance teachers can adapt to their students, subjects, and settings.