What to Do When a Project Goes Wrong
Science fair season may be winding down at most schools, but scientific exploration at home and in the classroom continues year-round. And where there is science, there are variables and materials and controls and reactions and things that change and bond and grow ... and things that don't.
Lots of things can go wrong with a project, even with a well-designed, well-scheduled, and conscientiously-run project.
Learning to handle a project that doesn't turn out exactly as expected and either regroup and get it back on track if there is time or deal with the unexpected results if the due date is too close for a repeat set of trials is important for students who are running scientific experiments of all sizes. It can be very frustrating when things go wrong. It can also be confusing, especially when you thought that you had done everything "right."
So what went wrong?
And what can you do about it?
What to Do Next
Our Staff Scientists have pooled their thoughts on troubleshooting a science project that doesn't work to help you step back, evaluate what happened, and figure out what you should do next.
- Are You Sure it Didn't Work?
It is important to first stop and ask yourself "How do I know my project 'failed'?"
"Maybe the problem is obvious," says Sandra, "like when you're putting together a circuit, and the light bulb at the other end fails to turn on. Or maybe you're re-creating a classic experiment like Gallileo's fabled Leaning Tower of Pisa Experiment, and you know what the proven hypothesis is," so you know it should have worked.
It gets trickier when you are working on an experiment of your own design. The question you have to ask yourself, says Sandra, is "did my project 'fail,' or was my hypothesis just incorrect?"
While student scientists can be disheartened if their initial educated guess turns out wrong, "proving a hypothesis wrong isn't bad science," reminds Sandra. For a student continuing research on the same topic, a failed hypothesis provides the groundwork for conducting further experiments to figure out why the initial guess was wrong.
- Before You Dive Back In...
It can be tempting to jump right back in, change things, add something here, remove something there. But the best approach is to step back and take some time to think through what happened before you begin troubleshooting—and before you repeat your experiment.
Our team all agrees it is important to take a deep breath and think about the project and the problem before you do anything hasty:
Kristin: "Ask yourself what you expected to see happen (what was the output that you anticipated) and what you saw happen."
Dave: "Be calm. Many procedures do not work flawlessly (or at all) the first time."
Michelle: "If a project doesn't work right away, don't start changing things willy nilly. Leave the project alone for a few hours and let your mind work things out."
- Review the Science Behind the Project
Doing some additional research, and re-reading any background materials that accompanied the project or procedures, can be an important step in troubleshooting. As Kristin explains, you want "to make sure that you understand the 'science' behind the experiment and what you expected to happen" so that you can effectively evaluate your results and analyze the procedures you used before trying again.
- Back to the Source
Re-read the full set of directions or the steps of the Experimental Procedure. Why? You may have overlooked a step that made all the difference between success and failure.
"As you read each step," says Sandra, "go back through your lab notebook and your memory and ask yourself: 'Did I do it exactly this way, or did I change this step in some way?'"
Any changes you made, or steps you forgot, are good bets for where things went wrong!
As you read through the project again, you'll want to pay special attention to the following:
- The Materials
One of the first things to doublecheck is the list of materials and supplies to ensure you used exactly what the project specified. Why? The wrong material could dramatically alter the outcome of an experiment. Similarly, if you knowingly made a substitution, even if you thought it would work, the changed material may have caused an unexpected result.
Kristin suggests that you not only look again at the materials list yourself, but enlist a friend, parent, or teacher to carefully go through the materials list (and procedure) with you. "There may be times when you misinterpret how to do something or miss a detail about a brand, size, or setup," she says.
Having an extra set of eyes look over the documentation with you can be really helpful.
- Evaluate Your Variables
Once you've reviewed the overall steps of your experiment, look carefully at your variables. Why? If you didn't treat your variables as outlined in the procedure, your results could certainly differ from the expected outcome. Maybe you misread something. Or, maybe you tried to take a shortcut?
You want to ask yourself two important questions, says Kristin: "What are my inputs (what am I changing)?" and "Did I change them as directed?"
- Evaluate Your Controls
A project that has built-in "controls" or "checkpoints" gives you clear points throughout the project where you can stop and evaluate your work or progress to make sure everything is right "at that point." Going back and looking at your results and progress at each control or checkpoint is an important step in figuring out what went wrong.
If the experimental procedure identifies the controls, you want to ask yourself: "Did a control fail?" It is possible you can you use the controls to pinpoint which step went wrong.
- When There are No Controls
Not all projects use controls. Sandra suggests that if you are working on a project that doesn't use controls, you may want to determine places where they can be added before you run your test again.
"Remember that it could be either a procedural step which is wrong, or some piece of equipment or material which is malfunctioning," says Sandra. "So you'll want controls which test as many of those things as possible."
Projects that involve building something may not yield traditional controls, so it's helpful to think about inserting "checkpoints" or steps where you could do or observe something that will indicate if everything is working "so far." For example, if you are working on building a complicated circuit, taking a reading with a multimeter at a certain point can indicate whether or not you are on the right track.
- The Procedure
Carefully re-reading the procedure, step by step and line by line, is a critical aspect in troubleshooting a project. These tips from our scientists can help as you review:
- Kristin: Pay close attention to any "notes" or images in the procedure that give clues about what you might observe, how the setup should look, or how you should conduct your testing.
- David: If there is a device that you have made, double-check any diagrams provided to make sure you assembled it correctly.
- The Materials
- Talk it Over
Talking over a "failed" project with a teacher or other adult can often be a good idea either before or after you work through the troubleshooting steps above. Sometimes, when you put things into words out loud, you'll hear the problem differently than when you are thinking it through on your own or on paper.
"Science doesn't happen in a vacuum," reminds Sandra. "Scientists talk to each other, and their collective experiences (or sometimes just the act of saying it all out loud) can spark the critical 'ah-ha' moment of understanding what went wrong or what needs to happen."
In the end, not all experiments will "work." If you're following someone else's Experimental Procedure (like a Science Buddies Project Idea), then you can probably feel confident that the project should work. Hopefully, careful troubleshooting using the guidelines and suggestions above will help you find the weak spot in the experiment you performed so that you can correct any problems and try again. But if your experiment was of your own design, it's possible it simply won't work as you've envisioned it this time around.
Troubleshooting can help you find what may be flaws in your design, and a review of the science behind the project and your ultimate goal for the project can help you shape and refine the procedure for subsequent trials and testing.
Don't lose heart.
Our scientists are quick to point out that not all science experiments "work" the first time.
"If your experiment 'failed,' consider yourself in good company," says Sandra. "The idea for many Nobel Prize-worthy science explorations started with someone scratching his or her head over a 'failed' experiment."
"Remember that negative results are real and important science, too," adds Kristin.
It's a good perspective to keep in mind. In fact, a "failed" experiment can be a stepping stone, says Sandra. "Understanding why an experiment failed can often lead you to a much more interesting, and unexpected, discovery."
Michelle agrees. "It is okay to fail. Remember what Thomas Edison said: 'Genius is 1% inspiration and 99% perspiration.' It took Edison many, many tries to find the right filament for the light bulb. Just keep working, and you will be able to figure things out."
Related Blog Posts: