Project-based learning: what it is, how to plan a unit, and what the research says
Teaching a unit through an extended project: students work over weeks on a real question or problem and make something for an audience, learning the content and skills along the way.
What the research says
- Strong
- Promising
- Mixed
- Thin
- Contradicted
Mixed, with an encouraging recent trend. Two cluster-randomized trials in US elementary schools found gains from carefully designed project units with teacher training: third graders in project-based science scored 0.277 standard deviations higher on a standardized science test, and second graders in project-based social studies grew more in social studies and informational reading. A trial in English secondary schools found no clear effect on literacy, though its evaluators rate that result low security. The evidence favors well-designed units with training over teachers building projects from scratch.
Rated Mixed. Two school trials support it. One rigorous study found no effect or the opposite. One review or weaker study adds context but doesn't move the rating. How we rate the evidence
Checked against each source in October 2026.
The studies behind this rating
Third graders taught four project-based science units, with materials and professional learning for teachers, scored higher on a standardized science test than the control group and reported more self-reflection and collaboration in science.
Sample 46 Michigan schools, 2,371 third graders, cluster randomizedWho School-age childrenEffect 0.277 standard deviation on a standardized science testRead it at the source (opens in a new tab)Second graders whose teachers taught four project-based social studies units, with professional development, grew more in social studies and informational reading than the comparison group, but not in writing or motivation. Classes that followed the unit plans more closely grew more in writing, motivation and reading.
Sample 48 second-grade teachers in low-income districts, randomly assigned in pairs within schoolsWho School-age childrenRead it at the source (opens in a new tab)A year of project-based learning for first-year secondary pupils had no clear impact on literacy or on engagement with school. There was a possible negative effect for pupils eligible for free school meals, which the evaluators treat with caution. Teachers and schools found it worthwhile, and it needed substantial management support.
Sample 24 English secondary schools (12 intervention, 12 control), Year 7 pupilsWho School-age childrenRead it at the source (opens in a new tab)A reply to Kirschner, Sweller and Clark: problem-based and inquiry learning are highly scaffolded, so they shouldn't be grouped with unguided discovery. The authors present evidence that both are effective, use scaffolding to reduce cognitive load, and serve goals such as collaboration and self-directed learning as well as content knowledge.
What it is
In project-based learning (often shortened to PBL or PjBL), the project isn't the dessert at the end of a unit; it's the unit. Students start with a driving question or a real problem ("How can we make our playground safer?", "What would help the birds that visit our school?"), and they learn the content and skills they need to answer it over several weeks. The unit ends with something real: a product, a presentation, a proposal, shared with an audience beyond the teacher.
The features most models share:
- A driving question that frames the unit and can't be answered with a quick search.
- Sustained inquiry over weeks, not a single lesson.
- Content taught along the way, often explicitly, when students need it for the project.
- Student voice and choice within a structure the teacher designs.
- Feedback and revision of drafts before the final product.
- A public product for an audience: families, another class, local experts.
It isn't the same as "doing projects". A diorama at the end of a unit is a project; it isn't project-based learning, because the learning happened before it, through other teaching. It also isn't unguided discovery: the successful programs in the research were carefully designed units with lessons, materials and teacher training.
Why it works
The case for it is that knowledge learned while doing something meaningful is more motivating, and more likely to be used again, than knowledge learned for a test. A project also gives a reason to read, write, calculate and talk.
Two recent randomized trials in US elementary schools give the best evidence. Joseph Krajcik and colleagues tested a set of four project-based science units for third grade, with materials and professional learning for teachers, in 46 Michigan schools with 2,371 students. Students in the project units scored higher on a standardized science test than the control group, and reported more self-reflection and collaboration in science.
Nell Duke and colleagues tested four project-based social studies units for second graders in low-income districts, with professional development, assigning 48 teachers at random. Students grew more in social studies and in informational reading, though not in writing or motivation. Classes that followed the unit plans more closely did better on writing, motivation and reading, a sign that how a project is taught matters.
A trial in English secondary schools, evaluated for the Education Endowment Foundation by Victoria Menzies and colleagues, found no clear impact of a year of project-based learning on literacy or engagement for first-year secondary pupils, and a possible negative effect for disadvantaged pupils that the evaluators treat with caution. The evaluators rate their own findings low security, mainly because many pupils were missing from the final analysis and some control schools adopted similar approaches. Teachers there found the approach worthwhile, and it needed a lot of support from school leaders.
How to start, by grade
Grades K–2
- Start with one short project (two to three weeks) tied to a science or social studies unit you already teach.
- Write a simple driving question in kid language: "How can we help new kids find their way around our school?"
- Plan the lessons first: the vocabulary, reading, writing and content children need, taught explicitly along the way.
- Use read-alouds of informational books as the main research source; young children can't yet read most of what they need.
- End with a simple product for a real audience (a map for kindergarteners, a short presentation for families), with one round of feedback and revision.
Grades 3–5
- Use a published, tested unit if one fits your standards, or adapt one rather than designing from nothing.
- Map each standard the unit must cover to a lesson inside the project, so content isn't left to chance.
- Give students roles and structured group tasks, with individual accountability for the learning.
- Build in checkpoints: a plan, a first draft, peer feedback, a revision. Use a rubric students see from the start.
- Bring in an outside audience or expert, even by video call, to make the product matter.
Grades 6–8
- Choose a driving question with real tension: a local issue, a design challenge, a decision with tradeoffs.
- Teach the hard content directly when students need it, in mini-lessons, rather than expecting them to discover it.
- Track progress with a project calendar and short individual checks so no one hides behind the group.
- Grade individual learning separately from the group product.
- Reflect at the end: what did we learn, and what would we change?
Grades 9–12
- Use projects for applying and extending knowledge students already have, not as the only way they meet new material.
- Partner with local organizations for authentic problems and audiences.
- Make the assessment criteria explicit and include an individual component such as a written report or oral defense.
Examples
Our school's birdsGrades 1–3
Driving question: "What would help the birds that visit our school?" Over three weeks, students learn about birds' needs through read-alouds and observation logs, count birds at a feeder, read and write short informational texts, and build simple feeders. They present a plan to the principal. Science content, reading, writing and counting all happen inside the project.
A local history guideGrades 2–4
Students research a place in their town using informational texts, photos and an interview with someone who remembers it. Each group writes and illustrates one page of a class guidebook, revises it after feedback from another class, and the finished guide goes to the town library. Social studies standards are taught in lessons along the way.
Designing a better recessGrades 4–6
Students survey classmates, graph the results, research the space and costs, and design a proposal for new recess equipment or games within a set budget. They present to the student council or principal. Math (data, area, money), writing and speaking are assessed individually as well as in the final proposal.
Water quality investigationGrades 6–8
Driving question: "Is the water in our local creek healthy?" Students learn the science of water quality in mini-lessons, collect and test samples, analyze data and write a report for a local environmental group, which visits to hear their findings.
Common mistakes
- Projects without explicit teaching.
- Students can't discover the water cycle or how to write an argument on their own. Plan the lessons inside the project and teach the content directly when it's needed.
- Losing the standards.
- A fun project that skips half the unit's content is a loss. Map every standard to a lesson or task inside the project before you start.
- Group products that hide individual learning.
- One student often does the work. Build in individual checks, individual writing and roles, and grade individual learning separately.
- Building every project from scratch.
- The positive trials used ready-made, tested units with training. If one exists for your standards, start there and adapt it.
- No feedback before the final product.
- Revision is where much of the learning happens. Schedule drafts and feedback, and give students time to act on it.
- Too big, too soon.
- A first project of two or three weeks with one subject is enough. Grow from there.
Time and cost
Time: several weeks per unit, usually replacing the existing unit rather than adding to it. Planning a project well takes far more time than a conventional unit, which is the main reason to start from a published one.
Money: modest for materials; more for published curricula and training.
Training: significant. Both positive trials included professional development for teachers, and the English trial's evaluators found the approach needed substantial management support and changes to how the school was organized.
Debates and open questions
Is PBL the same as discovery learning? Critics have grouped project-based and inquiry approaches with unguided discovery, which does poorly in the research (see the explicit instruction card). Cindy Hmelo-Silver and colleagues replied that well-run problem- and inquiry-based learning is heavily scaffolded. The successful trials here support that reading: they were structured units with teaching built in.
What about older students? The two positive trials are in elementary grades. The secondary trial found no clear effect, with low security. Evidence for middle and high school is less clear from the studies we cite.
Motivation. It's often claimed that projects motivate students. Duke and colleagues measured motivation and found no overall difference, though classes that stuck closely to the unit plans did better. Krajcik and colleagues found more self-reflection and collaboration in science. So the motivation benefit is plausible but not guaranteed.
Equity. The English trial found a possible negative effect for disadvantaged pupils, which its evaluators treat with caution; the Duke trial was run in low-income districts and found gains. Structure and support for students who arrive with less background knowledge seem to matter.
Tools that do it
Free, on this site, nothing to sign up for. Class lists stay in your browser.
- Rubric maker: a rubric for the final product and the process, shared with students from day one.
- Random group maker: form project groups, or keep the same ones for the whole unit.
- School calendar maker: a project calendar with checkpoints and the presentation date.
- KWL chart: what we know, what we want to know, what we learned: a simple project launch and close.
- Unit planning page: map the standards to lessons inside the project before you start.
- Countdown timer: keep work sessions focused and on schedule.
Questions teachers ask
- What's the difference between project-based learning and doing projects?
- When the project comes after the learning (a poster or diorama to show what you learned), that's doing projects. In project-based learning the project is the unit: students learn the content and skills while working on it over several weeks, toward a product for a real audience.
- Does project-based learning work?
- In the best recent trials, yes, when teachers used well-designed units with training: third graders gained in science and second graders gained in social studies and informational reading. A secondary-school trial in England found no clear effect, with low security. The rating is Mixed for that reason.
- How long should a project be?
- Start with two or three weeks for a first project, tied to one unit. Each of the two positive trials used a set of four units; build up to that as you and your students get used to the structure.
- Do I still teach lessons in PBL?
- Yes. The successful programs had lessons built into the units. Teach new content explicitly when students need it, then let them use it in the project.
- How do I grade a group project fairly?
- Separate the group product from individual learning. Use a rubric for the product, and individual checks (a short quiz, a written reflection, an individual section) for each student's learning.
- Is PBL good for struggling students?
- It can be, with structure. The second-grade trial was in low-income districts and found gains. The English trial raised a possible concern for disadvantaged pupils, which its evaluators treat with caution. Explicit teaching inside the project and clear support make the difference.
Related in the Library
Cite this card
Written for teachers by Peabody Teach from the sources listed above, which we read on the publisher's or clearinghouse's own page. We summarize and link; we don't copy. No ads, no affiliate links. Something wrong or out of date? Tell us.