Table of Contents
STEM vs STEAM Education: What the Research Actually Shows
Walk through a school in 2026 and you're likely to find the word STEAM on a bulletin board, a classroom door, or a program flyer. The acronym — Science,.
STEM vs STEAM Education: What the Research Actually Shows
Walk through a school in 2026 and you’re likely to find the word STEAM on a bulletin board, a classroom door, or a program flyer. The acronym — Science, Technology, Engineering, Arts, Mathematics — has become nearly universal in education marketing. Schools add it to curriculum descriptions; summer programs use it to signal creativity and breadth; grant applications invoke it to suggest holistic learning. The question that gets asked less often is whether adding an A to STEM actually changes outcomes for kids, and if so, under what conditions.
This is a question the research can partially answer. The partial answer is more useful than the marketing materials suggest, but more limited than the strongest STEAM advocates claim.
Key Takeaways
- Specific, well-designed arts integration in STEM contexts has documented benefits — particularly music instruction and spatial reasoning, visual arts and data representation, and design thinking as an engineering methodology.
- Broad STEAM branding with weak arts integration — adding craft activities to science units, calling existing maker projects “STEAM” — shows little evidence of benefit beyond what the STEM components would produce alone.
- The evidence base is genuinely thin for universal STEAM claims, partly because “arts integration” describes an enormous range of practices, from deeply embedded curriculum design to superficial add-ons.
- Music and spatial reasoning is the most robust specific connection, with multiple studies showing music training improving spatial visualization skills that support mathematics and engineering.
- Implementation quality — how deeply the arts are integrated with the specific STEM content — matters more than whether a program has STEAM in its name.
The Problem: A Popular Concept With Uneven Evidence
STEM education emerged as a policy priority in the 2000s, driven by labor market demand projections and comparative international assessment data. The framework was explicit: improve science, technology, engineering, and mathematics education because those are the fields driving economic growth and those are the assessments on which U.S. students were underperforming relative to international peers.
The arts were not in that frame. STEAM emerged partly as a corrective — the argument being that STEM without creative thinking produces engineers who can solve specified problems but can’t frame new ones, scientists who can run experiments but can’t communicate findings to a broader public, and technologists who can code but can’t design for human experience. The STEAM movement, associated particularly with Rhode Island School of Design (RISD) and its advocacy in the early 2010s, argued that arts and design thinking were not supplementary to STEM but integral to innovation.
This argument has intuitive appeal. Apple’s design under Steve Jobs, the aesthetic quality of well-engineered consumer products, the role of information visualization in scientific communication — these are real phenomena. The question for education research is whether integrating arts into K-12 STEM curricula actually improves STEM outcomes, and what kind of integration does so.
The honest answer from the research literature is: it depends almost entirely on what you mean by “arts integration.” The term covers everything from music theory to visual arts to drama to dance to craft-based making activities. These are not the same thing, they don’t share the same cognitive mechanisms, and they don’t produce the same effects on STEM outcomes. Evaluating STEAM as a unified concept is like evaluating “sports” as a health intervention — the average tells you almost nothing useful.
What the Research Actually Says
Winner, Goldstein, and Vincent-Lancrin (2013) — “Art for Art’s Sake? The Impact of Arts Education.” This OECD report by Ellen Winner and colleagues is one of the most comprehensive and methodologically careful reviews of the arts education literature. The title’s irony is deliberate: the authors were skeptical of broad claims for arts education transfer effects. Their review found that the evidence for arts instruction improving academic outcomes in non-arts domains is much weaker than arts advocates typically claim. They found some evidence for specific transfer effects — music and spatial reasoning, visual arts and attention to detail, theater and verbal skills — but no support for the broad claim that general arts education improves academic outcomes across domains. They called for abandoning the claim that arts education produces universal academic benefits and instead focusing on documenting specific mechanisms where transfer is real and robust.
Hetland and Winner (2001) — Music and spatial reasoning. Lois Hetland and Ellen Winner conducted a meta-analysis of studies examining the relationship between music instruction and spatial reasoning. This is the most reliable specific connection in the arts-STEM evidence base. Their analysis found a consistent effect: music instruction, particularly keyboard instruction with training in reading musical notation, improves spatial-temporal reasoning — the ability to mentally transform and rotate objects in space, which is directly relevant to geometry, engineering design, and scientific visualization. The effect sizes were meaningful (d ≈ 0.35–0.57 across studies) and replicated across multiple independent research groups. The mechanism is reasonably well understood: music reading and production train the neural systems involved in spatial-temporal processing.
LaJevic (2013) — Arts integration in STEM: Examining teacher practice. Lisa LaJevic’s qualitative research on arts integration in elementary STEM classrooms documented a pattern that appears throughout the STEAM literature: teachers who deeply understand both the arts and the STEM content produce genuine integration, while teachers who don’t have this dual expertise tend to use arts activities as decoration or reward rather than as instructional tools. A science lesson where students “draw what they observed” is different from one where visual representation is used as a scientific tool — a way of making observations more precise, noticing features they might otherwise miss, or communicating findings to others. The drawing is similar; the cognitive demand and learning function are completely different. LaJevic’s work suggests that STEAM outcomes depend more on teacher preparation and depth of content knowledge than on curriculum labels.
Hardiman, Rinne, and Yarmolinskaya (2014) — Arts integration and memory. This experimental study at Johns Hopkins tested whether learning science content through arts integration (combining science concepts with visual art and drama-based activities) improved memory for the content. Students in arts-integrated lessons showed better long-term retention of science content than students in equivalent lessons without arts components, even when the time-on-task was equivalent. The proposed mechanism is that arts activities — particularly dramatic enactment and visual creation — engage multiple encoding pathways (verbal, visual, kinesthetic) simultaneously, which produces more durable memory traces. This is a modest but real effect, and it suggests that arts integration can improve STEM learning at the content retention level even when it doesn’t change higher-order STEM skills.
Peppler and Wohlwend (2018) — Making and STEAM: Distinguishing craft from inquiry. Kylie Peppler and Karen Wohlwend’s work distinguishes between “making” activities that are primarily craft (following directions to produce a predetermined product) and making activities that involve genuine design inquiry (defining a problem, generating multiple solutions, testing, iterating). Many programs labeled STEAM are primarily the former — students build specific things following specific instructions, incorporating visual design elements. This produces engagement and can develop fine motor skills and attention to craft, but it doesn’t develop the design-thinking or engineering-design capacities that STEAM advocates typically cite as the rationale for arts integration. Genuine design inquiry, where students define the problem and constraints themselves and iterate based on testing, shows more robust connections to engineering design competencies.
| Arts Integration Type | Specific STEM Benefit | Effect Size / Evidence Quality | Conditions Required |
|---|---|---|---|
| Music instruction (keyboard, notation) | Spatial reasoning, geometric visualization | d ≈ 0.35–0.57; multiple replications | Sustained instruction (months, not sessions); notation reading required |
| Visual arts as scientific observation tool | Attention to detail, precision in observation | Moderate; documented in biology, natural science | Teacher must frame drawing as inquiry, not decoration |
| Drama / embodied enactment of science content | Content memory retention | Small-moderate; single strong study (Hardiman et al.) | Enactment must represent the mechanism, not just the topic |
| Design thinking as engineering methodology | Problem framing, iterative design, solution generation | Moderate; documented in design research, weaker in STEM assessment outcomes | Students must define problems and constraints; open-ended, not pre-specified |
| Visual arts / data visualization | Quantitative communication, pattern recognition | Emerging; some evidence in data science education | Must link art decisions to data meaning, not purely aesthetic |
| Craft-based making (directed) | Engagement, fine motor skills, following instructions | Small academic benefit; consistent engagement effect | Insufficient for STEM skill development claims |
| Dance / movement for math concepts | Limited; primarily used in early childhood | Very limited evidence for sustained academic benefit | — |
| General “STEAM” exposure (mixed activities) | Inconsistent; depends entirely on implementation | Cannot be evaluated as unified category | Quality of integration; teacher expertise in both domains |
What to Actually Do
Understanding the specific evidence helps parents evaluate programs more effectively than STEAM/STEM labels do.
Ask About Depth of Integration, Not Breadth of Activity
A program that does art one day and science another, calling the combination STEAM, is doing something different from a program where artistic practice is embedded in the scientific or engineering process — where students are using visual representation as a tool for scientific thinking, or where design constraints require both aesthetic and functional judgment. Ask programs to explain how the arts component is integrated with specific STEM learning goals. If the answer is “we do science projects and also have an art class,” that’s not the integration the research supports.
Take Music Training Seriously as a STEM-Supporting Activity
The music-spatial reasoning connection is the most replicated specific finding in the STEAM literature. If your child is interested in music and you’re wondering whether music lessons “count” toward STEM development, the answer based on the evidence is: keyboard instruction with music reading does appear to support spatial reasoning in ways relevant to mathematics and engineering. This isn’t a reason to force unwilling children into piano lessons, but it’s a reason not to see music as categorically separate from STEM development.
Evaluate Design Thinking Claims Carefully
“Design thinking” has become as overused a term as STEAM itself. True design thinking — as described in engineering design frameworks and documented in research — involves open-ended problem framing (not just problem-solving), constraint identification, multiple solution generation, prototyping, testing, and iteration based on results. A program where students follow step-by-step instructions to build a specific thing and then decorate it is not doing design thinking. Ask: do students define the problem? Do they generate multiple different solutions? Do they test and change their design based on results?
Don’t Dismiss Arts Education Because STEAM Claims Are Overstated
Winner et al.’s skeptical review is sometimes misread as arguing that arts education has no value. That’s not their argument. Their argument is that arts education should be justified on its own terms — that learning visual art, music, theater, and design are valuable in themselves — rather than justified by inflated claims about transfer to non-arts academic domains. Art for art’s sake is a legitimate educational goal that doesn’t need STEM transfer effects to justify it. Parents and educators who reject arts education because they’ve heard the transfer claims are overstated have misread the research.
For Home: Use Drawing as a Scientific Tool, Not a Reward
One of the most accessible and evidence-supported arts-STEM integrations for home use is scientific drawing — using detailed observational drawing as a tool for careful observation, not as an art activity added to a science activity. When children draw a plant from observation, an insect they found, the cross-section of a fruit, or the progression of a cloud formation, the drawing demands a quality of attention that looking alone doesn’t require. Have your child draw first, then look up photos or information. The comparison between what they drew and the reference image becomes a scientific conversation about what they noticed and what they missed.
What to Watch for Over the Next 3 Months
Week 4: If your child is in a STEAM-labeled program, ask them to explain one specific way they used art as a tool for their science or engineering work — not an art project they did alongside science, but a moment where making or looking at something visual helped them understand or communicate something about the STEM content. If they can’t describe one, the integration may be superficial.
Month 2: For children taking music lessons: have you noticed any transfer to math or spatial tasks? This won’t be visible in test scores at this time scale, but you might notice it in puzzle-solving, map-reading, geometry homework, or 3D visualization tasks. The transfer from music to spatial reasoning is documented on the timescale of months of sustained instruction, not weeks.
Month 3: For programs that use design thinking: has your child worked on a project where they changed their design based on a test result? Iteration driven by feedback is the core mechanism of design thinking. If every project ends with the first design, the design thinking component is more rhetorical than instructional.
Frequently Asked Questions
Does STEAM education actually improve test scores?
The honest answer is: not reliably, and not in the short term. The studies that show academic benefits from arts integration mostly show effects on content retention and specific spatial skills, not on broad academic achievement tests. Programs that market STEAM as a test-score improvement strategy are overstating the evidence.
Is there any evidence that STEAM approaches improve STEM career interest?
There’s some evidence that arts integration improves engagement and enjoyment in STEM activities, particularly for students who might otherwise find STEM alienating. Whether this translates to sustained career interest is much harder to demonstrate. The most credible mechanism is that arts-integrated STEM reduces the perception that STEM is purely analytical and humorless — which may matter particularly for students whose identity doesn’t fit the “typical STEM student” stereotype.
My child’s school does STEAM but it looks like science and art alternating. Is that useful?
Alternating science and art classes is not STEAM integration in the research sense — it’s just a broad curriculum that includes both. That’s fine; both science and art are worth teaching. But the documented benefits of arts integration apply to practices where the two are genuinely connected — where artistic skill is deployed as a tool within scientific or engineering inquiry. Separate classes doing separate things is just two separate things.
Does the order matter — should kids do the art first or the science first?
For integrated STEAM activities, the sequence should be determined by the learning goal. If the goal is to use artistic observation to develop scientific noticing skills, observation drawing should precede reading about the subject. If the goal is to communicate scientific findings visually, the science comes first. The integrated design is what matters, not a canonical sequence.
How do I evaluate whether a STEAM camp or program is actually doing what it claims?
Ask to see a sample project or lesson and trace the connection: at what specific point does the artistic skill or artistic thinking directly serve the STEM learning objective? If the counselor can explain that clearly, the program is doing something real. If the answer is “we do both,” it’s probably just breadth rather than integration.
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
- Winner, E., Goldstein, T., & Vincent-Lancrin, S. (2013). Art for art’s sake? The impact of arts education. OECD Publishing.
- Hetland, L., & Winner, E. (2001). The arts and academic achievement: What the evidence shows. Arts Education Policy Review, 102(5), 3–6.
- LaJevic, L. (2013). Arts integration: What is really happening in the elementary classroom? Journal for Learning Through the Arts, 9(1).
- Hardiman, M., Rinne, L., & Yarmolinskaya, J. (2014). The effects of arts integration on long-term retention of academic content. Mind, Brain, and Education, 8(3), 144–148.
- Peppler, K., & Wohlwend, K. (2018). Theorizing the nexus of STEAM practice. Arts Education Policy Review, 119(2), 88–99.
- Hetland, L., Winner, E., Veenema, S., & Sheridan, K. M. (2007). Studio thinking: The real benefits of visual arts education. Teachers College Press.
- Catterall, J. S. (2009). Doing well and doing good by doing art. ImageNation.
- Ruppert, S. S. (2006). Critical evidence: How the ARTS benefit student achievement. National Assembly of State Arts Agencies.
