Why the Best STEM Learning Happens at the Kitchen Table, Not on a Screen

The app has a name like “Science Lab Pro” or “STEM Explorer” and costs $8 a month. It has star ratings and testimonials and a curriculum map. Your child plays it for three days, finds the secret to unlocking everything, and moves on.

Meanwhile, the vinegar-and-baking-soda experiment your child did on a Tuesday because they were bored — where they kept changing the amounts to see what would happen, where they asked why it fizzed and you said “I think it’s a gas, what do you think?” and they went looking for an answer — that was more science than three months of the app.

This isn’t an accident. The research on informal science learning has been building for twenty years, and the picture is increasingly consistent: the home learning environment — with real objects, genuine curiosity, and adults who ask rather than tell — produces stronger STEM identity and long-term science engagement than most packaged educational technology. Household-object science has more evidence behind it than most EdTech apps do.

What “Informal Science Learning” Means and Why Researchers Are Excited About It

Formal science education is school: curriculum, textbooks, standardized tests, scheduled instruction. Informal science learning is everything else — museums, science centers, after-school programs, weekend experiments at home, conversations in the car.

The National Research Council’s landmark 2009 report Learning Science in Informal Environments, led by Philip Bell and colleagues, synthesized decades of research and identified six learning strands that informal science environments uniquely support: experiencing excitement and interest, understanding science knowledge, engaging in scientific reasoning, reflecting on science as a way of knowing, participating in science practices, and developing science identity.

That last one — science identity — turns out to be enormously important. Science identity is the degree to which a child sees themselves as a “science person.” Research consistently shows it predicts long-term STEM participation more reliably than test scores do. A child who scores 90% on a science test but doesn’t identify as a science person is less likely to pursue science than a child who scores 75% but genuinely thinks of themselves as someone who does science.

John Falk and Lynn Dierking’s contextual model of learning (2000) identifies the home as the most influential context for science identity development — more than school, more than museums, because the home provides the longest exposure with the most personally meaningful objects and relationships.

The implication is not that school doesn’t matter. It’s that what happens in informal contexts — especially the home — shapes whether formal STEM education lands on prepared or unprepared ground.

The Home Is Already a Science Lab — You’re Just Not Using It That Way

The kitchen alone contains more science learning potential than most elementary school labs. Consider what’s already there:

Baking is applied chemistry, thermodynamics, and materials science. The Maillard reaction (why bread browns), the physics of yeast, the protein chemistry of gluten — these are university-level science concepts that a 9-year-old can engage with at the “why does this happen” level.

The freezer demonstrates phase transitions, density changes, and pressure differentials. Why does ice cream get ice crystals when you put it back? Why does water expand when it freezes? Why is frost on the inside of the freezer, not the outside?

The sink and drain illustrate fluid dynamics, viscosity, and gravity. Change the angle of a cup pouring water. Watch how soap disrupts water’s surface tension (which is why it cleans). Ask why hot water drains faster than cold.

Plants on the windowsill are photosynthesis, cellular respiration, and growth responses (phototropism, gravitropism). Change which direction the plant faces. See what happens in a week.

None of this requires buying anything. It requires a disposition of curiosity and adults who treat questions as worth investigating rather than dismissing.

The Afterschool Alliance’s 2014 national report found that children who report high levels of science-related activity at home — including conversations about science, experiments, and visits to science-related venues — are significantly more likely to aspire to STEM careers, control for all other factors.

Informal Learning vs. Formal EdTech: STEM Identity Outcomes

The evidence comparing informal home learning to EdTech applications is limited in direct comparisons but suggestive. No large-scale randomized study has pit “kitchen experiments” against “educational apps” for STEM identity. What exists is a strong body of evidence on each separately.

Learning Approach	Cost	Prep Time	Research Support	STEM Identity Impact	Completion/Engagement
Household object experiments (vinegar+baking soda, magnets, etc.)	Near zero	Low (5–10 min)	Strong (NRC 2009, Falk & Dierking)	High — strongest predictor of science identity	High — child-directed, no deadline
Science kits (pre-packaged)	$25–$80/kit	Low	Moderate	Moderate — depends on adult involvement	Variable — novelty-driven
Screen-based science apps	$0–$10/mo	None	Weak — most apps have no independent efficacy research	Low to moderate	Low — median engagement ~3 days
Educational science videos (YouTube, etc.)	$0	None	Weak for skill outcomes	Low to moderate (motivational)	Very low for sustained learning
In-person science museums	$15–$25/visit	Moderate	Strong (Bell et al. 2009)	High — especially with adult mediation	High during visit
After-school STEM programs	$100–$400/mo	None for parent	Strong (RAND, Afterschool Alliance)	High — especially with project completion	Moderate–high

The pattern: household-based physical activities and after-school programs have the strongest research support for STEM identity outcomes. Most apps lack independent efficacy data — meaning the evidence is testimonials, user engagement metrics, and curriculum alignment claims, not learning outcome research.

This doesn’t mean apps are useless. It means their evidence base is thin compared to what we know about informal science learning at home. The NRC report’s research had decades of rigorous study behind it. The app in the app store has a star rating.

8 Kitchen Table Activities With the Strongest Science Learning Evidence

These activities recur in informal science learning research because they produce the conditions that build science identity: genuine uncertainty, observable outcomes, and questions worth investigating.

1. Density Tower

Layer liquids of different densities (honey, corn syrup, dish soap, water, vegetable oil, rubbing alcohol with food coloring) in a clear glass. Each layer floats on the one below it. The questions cascade: why don’t they mix? What happens if you add an ice cube? What if you warm the glass? This is physics and chemistry at a level that is visually compelling and conceptually deep.

2. Non-Newtonian fluid (oobleck)

Mix cornstarch and water. Squeeze it hard and it acts like a solid; pour it slowly and it flows like liquid. This is a direct experience of a non-Newtonian fluid, which your child will not understand fully — and that’s the point. A 7-year-old walking away saying “it acts like a solid when you press hard and a liquid when you don’t, but I don’t know why” is doing more science than one completing a worksheet about states of matter.

3. Paper tower challenge

One sheet of paper. No tape, no glue. Fold it however you want. Make it stand as tall as possible. This is structural engineering: compression, tension, and moment arms at a kitchen-table scale. Adding constraints (it has to hold a coin at the top) introduces load-bearing design.

4. Growing mold on bread

Put bread slices in plastic bags with different moisture levels and temperatures. Observe and record over two weeks. This is biology and microbiology — growth conditions, hypothesis testing, controlled variables. The child is designing an experiment, not following one.

5. Egg in saltwater vs. freshwater

An egg sinks in fresh water. Add enough salt and it floats. The same principle governs why some people float more easily in the ocean than in a pool, and why the Dead Sea lets anyone float. This is density and buoyancy, and the variable (how much salt?) is fully controllable by the child.

6. Balloon static electricity

A balloon rubbed on hair picks up small pieces of paper, deflects a thin stream of water, and can make another balloon repel. This is electrostatics — charge transfer, electric fields, and force at a distance. Every phenomenon has a testable variation: does a dry balloon work better than a damp one? Does it work on different materials?

7. Cabbage juice pH indicator

Red cabbage juice changes color from purple (neutral) to pink/red (acidic) to green/yellow (basic). Test vinegar, baking soda solution, lemon juice, and tap water. This is real analytical chemistry — the same principle used in pH meters, just with cabbage instead of electronics.

8. Paper bridge load testing

Fold a sheet of paper into a bridge between two books. Add pennies one at a time until it collapses. Then change the fold geometry and repeat. This is structural testing — the scientific method applied to engineering. Documenting predictions before adding each penny introduces the habit of recording hypotheses.

How to Ask Questions That Turn Making Into Learning

The research on adult mediation in informal science is remarkably consistent: what parents say while a child experiments matters more than the experiment itself.

Bell and colleagues’ 2009 NRC report and subsequent studies identify elaborated conversation — asking “what do you think will happen if…?” and “why do you think that happened?” — as the primary mechanism by which informal science activities translate into science identity. Adults who provide answers short-circuit this process. Adults who ask questions extend it.

Three question types that research identifies as particularly valuable:

Prediction questions (“What do you think will happen?”): These require the child to form a mental model before the observation, which makes the observation informative regardless of whether the prediction was right.

Cause questions (“Why do you think that happened?”): These direct attention to mechanism — not just “it happened” but “why it happened.” Science is about mechanisms.

Extension questions (“What if we changed this?”): These push toward variable manipulation, which is the heart of experimental thinking.

What to avoid: explaining the answer. “That fizzes because the acid reacts with the base to produce carbon dioxide” is less educationally valuable than “what do you think is making the bubbles?” The former gives the child information. The latter engages their reasoning.

Signs Your Kid Is Building a Real Science Identity

These are observable and don’t require test scores.

They ask “why” about things that aren’t assigned. They notice phenomena in everyday life and wonder about mechanisms — why the car window fogs up from the inside, why the pasta water boils faster with a lid, why the bread rose more today than last time.

They propose their own experiments. “What if we tried it with ice instead of warm water?” is a child generating a hypothesis and proposing a test. This is the scientific method, spontaneously applied.

They tolerate uncertainty. A child with emerging science identity can say “I don’t know why that happened” without distress — and then try to find out, or accept that they don’t know yet. This is one of the most durable skills science education can produce.

They make predictions before observing. “I think the denser one will sink” before the demonstration, not after. Predicting before observing requires a mental model, which means learning happened.

These behaviors are what the research on building problem-solving skills through active making connects to long-term capability — and they’re what distinguishes home learning that produces STEM identity from screen-based learning that produces entertainment engagement.

For how this informal foundation connects to formal skill development, see also: free vs. paid STEM programs and what actually determines outcomes.

Key Takeaways

• The NRC’s 2009 Learning Science in Informal Environments report identified the home as a primary context for science identity development — stronger than school for identity specifically
• Science identity — whether a child sees themselves as a “science person” — predicts long-term STEM participation more reliably than test scores
• Most EdTech science apps have no independent learning outcome research; informal household science activities have decades of supporting evidence
• Adult questioning during experiments (“why do you think that happened?”) is more educationally valuable than adult explanation — and this effect is documented specifically in informal science learning research
• The eight activities above recur in research because they produce genuine uncertainty, observable outcomes, and questions worth investigating with simple, cheap, household materials
• STEM identity signals to watch for: child-generated “why” questions, spontaneous hypothesis generation, comfort with uncertainty, and predicting before observing

FAQ

How much time should we spend on kitchen table science each week?

Research on informal science learning doesn’t prescribe a specific time. Quality matters more than quantity. A 20-minute experiment where a child generates predictions, observes outcomes, and discusses why something happened produces more STEM identity benefit than three hours of passive educational video.

My child doesn’t seem interested in science at all. Will these activities help?

Possibly — particularly if the activities connect to existing interests. A child interested in cooking who learns why bread rises is encountering chemistry through a domain they care about. The research on science identity development suggests that interest-triggered engagement with science phenomena is one of the primary pathways to identity formation.

Are science kits better than household materials?

Kits have the advantage of providing materials and some structure, which reduces parent preparation. Research suggests outcomes are comparable to household-material experiments when adult mediation is similarly engaged. The advantage of household materials is that the child observes science happening in the context of everyday life, which strengthens the connection between science and the real world.

Do these activities replace formal science education?

No — and the research doesn’t suggest they should. The NRC report explicitly treats informal and formal science as complementary systems. Informal learning builds the identity and curiosity that formal education can build on. Children with strong informal science backgrounds tend to engage more productively with formal science instruction.

My child is easily frustrated when experiments don’t produce expected results. Is that a problem?

Frustration tolerance in the face of unexpected results is a learnable disposition, not a fixed trait. The experimental framing helps: “the experiment didn’t fail, it told us something unexpected — what did it tell us?” Shifting from failure to information reframes the unexpected outcome as data. This reframing is a cognitive habit that takes practice.

What’s the minimum viable kitchen science setup?

Vinegar, baking soda, food coloring, paper, tape, and a clear glass. That’s enough for at least 10 distinct experiments with genuine scientific content. Everything else is bonus.

---

About the author

Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.

Sources

1. Bell, P., Lewenstein, B., Shouse, A. W., & Feder, M. A. (Eds.). (2009). Learning Science in Informal Environments: People, Places, and Pursuits. National Research Council, National Academies Press. https://doi.org/10.17226/12190
2. Falk, J. H., & Dierking, L. D. (2000). Learning from Museums: Visitor Experiences and the Making of Meaning. AltaMira Press.
3. Afterschool Alliance. (2014). STEM Learning in Afterschool: An Analysis of Impact and Outcomes. https://www.afterschoolalliance.org/documents/STEM-Afterschool-Outcomes.pdf
4. National Science Foundation. (2023). Advancing Informal STEM Learning (AISL) Program. NSF.gov. https://www.nsf.gov/funding/pgm_summ.jsp?pims_id=504793
5. RAND Corporation. (2016). Afterschool Programs in the 21st Century. https://www.rand.org/pubs/research_briefs/RB9999.html
6. Falk, J. H., Storksdieck, M., & Dierking, L. D. (2007). “Investigating public science interest and understanding: Evidence for the importance of free-choice learning.” Public Understanding of Science, 16(4), 455–469. https://doi.org/10.1177/0963662506064240
7. National Research Council. (2012). A Framework for K–12 Science Education: Practices, Crosscutting Concepts, and Core Ideas. National Academies Press. https://doi.org/10.17226/13165