The first time I run a STEM challenge each year, a hand shoots up the moment a tower wobbles and falls: was that wrong? That single question is the whole unit waiting to be taught. My students walk in believing engineering means getting the answer right on the first try, and the engineering design process is built on exactly the opposite idea.
Engineering design is not a straight line from problem to solution. It is a loop: you define the problem, imagine solutions, build a prototype, test it, learn from what broke, and go around again. The NGSS engineering standards MS-ETS1-1 through MS-ETS1-4 reward that loop directly, and a good STEM challenge is the fastest way I know to make it real. Here is how I teach it.
What is the engineering design process?
The engineering design process is an iterative, cyclical set of steps engineers use to solve problems: define the problem with its criteria and constraints, brainstorm possible solutions, plan and build a prototype, test it and collect data, evaluate the results against the criteria, then improve the design and try again. It is not strictly linear. Engineers loop back and refine, again and again.
- Define the problem, including its criteria for success and its constraints.
- Brainstorm and imagine possible solutions.
- Plan and build a prototype.
- Test the prototype and collect data.
- Evaluate the results against the criteria.
- Improve the design and repeat the cycle.
What are criteria and constraints? (MS-ETS1-1)
Criteria are what a successful solution must do, and constraints are the limits it must work within, such as time, materials, or cost. MS-ETS1-1 asks students to define a design problem precisely enough that its success can be judged. If a class cannot name the criteria and constraints, they cannot tell whether a design actually worked, so I make students write them down before anyone builds.
I keep the two words straight with a simple frame: criteria are the goals, constraints are the limits. A bridge must hold a certain weight (criteria) using only paper and tape, finished in one class period (constraints). Pinning these down first is what turns a fun build into real engineering, because now there is a clear standard to measure every attempt against.
How is engineering design different from the scientific method?
Scientists investigate questions about how the natural world works; engineers solve problems and optimize solutions. The scientific method seeks an explanation, while the engineering design process seeks a working design that meets criteria within constraints. Both test ideas and rely on evidence, but their goals differ: one is asking why, the other is asking how to make something work better.
Students often assume the two are the same thing with different names, so I draw the line clearly. A scientist asks why a ball rolls farther on smooth floor. An engineer asks how to build a car that rolls the farthest. Same evidence, same testing, different goal. Knowing the difference helps students see that in a STEM challenge, their job is to optimize, not just to explain.
Why is failure part of the engineering design process? (MS-ETS1-2, MS-ETS1-3)
Failure is not the opposite of success in engineering, it is information. A failed test gives data about what to improve, which is the heart of MS-ETS1-3, analyzing test data to refine a design. MS-ETS1-2 asks students to evaluate competing solutions against the criteria. Both standards treat a design that breaks as a useful result, not a mistake to be embarrassed about.
When a prototype falls apart, I do not say bad luck, I say great, now we have data. I have students record what failed and why, then change one thing and test again. That reframe is the most important thing I teach all unit. Once a class stops fearing failure and starts mining it for clues, they iterate fast, and iteration is where the real engineering lives.
What STEM challenges teach the engineering design process? (MS-ETS1-4)
The best challenges force students to build, test, and improve a model, which is exactly what MS-ETS1-4 asks. Low-prep paper-and-tape builds get the whole class iterating quickly, a multi-day vehicle challenge lets them refine across several test runs, and a review activity locks in the vocabulary. Pick a challenge with clear criteria and real constraints, and the process teaches itself.
- Paper and tape builds: quick, low-material challenges that let every student run the full loop in a class period.
- A multi-day vehicle challenge: more test runs means more chances to analyze data and iterate toward a better design.
- A review escape room: scenario puzzles reinforce the steps, criteria, and constraints under a little friendly pressure.
Teach the loop instead of the right answer, treat every failed test as data, and a wobbling tower stops being a mistake and becomes the best teaching moment of the day.