Why Unexpected Results Can Be Useful

When I do an experiment, I usually have an idea of what I expect to happen.

I make a prediction, plan the experiment, and collect data.

But sometimes the results don't match my prediction at all.

At first, that can feel like a problem.

But unexpected results can actually be useful.

They can show scientists something they didn't notice before, reveal problems with an experiment, or lead to new questions. Scientists can learn from results that don't match their expectations instead of simply ignoring them. (rsc.org)

When Results Surprise You

Imagine I predict that plants receiving more sunlight will grow taller.

I run the experiment and collect my measurements.

But the plants receiving less sunlight actually grow taller.

That's unexpected.

I shouldn't immediately decide that the experiment was useless.

Instead, I should ask:

Why did this happen?

That question could lead me to investigate what went wrong or what I didn't understand.

Check the Experiment First

Before deciding that an unexpected result is a major discovery, scientists need to make sure the experiment itself was reliable.

I could check:

• Did I measure everything correctly?
• Did I follow the procedure?
• Did I accidentally change another variable?
• Was my equipment working properly?
• Did I record the data correctly?
• Did something unusual happen during the experiment?

Sometimes an unexpected result is caused by an error.

Other times, it is a real result that needs more investigation.

Unexpected Results Can Reveal Problems

Suppose I am measuring how quickly a chemical dissolves at different temperatures.

Most of my measurements make sense, but one result is completely different.

That could mean something went wrong.

Maybe I measured the temperature incorrectly.

Maybe I recorded the time incorrectly.

Maybe the sample was different from the others.

Finding an unusual result can help me discover a weakness in my experiment.

Checking data for surprising patterns is also a useful way to find possible mistakes. (journals.plos.org)

Sometimes the Result Is Real

Not every unexpected result is an error.

Sometimes an experiment produces a real result that the scientist didn't expect.

That can be especially interesting.

Scientists may repeat the experiment to see whether the same result happens again.

If the result keeps appearing, it becomes more interesting because it may represent a real pattern rather than a random mistake.

Unexpected experimental results can sometimes provide clues for further investigation. (pmc.ncbi.nlm.nih.gov)

Unexpected Results Can Create New Questions

One of my favorite things about science is that one question can create another.

Suppose I am testing whether more sunlight makes plants grow taller.

The results show something completely different.

Now I might ask:

Does too much sunlight actually slow plant growth?

Or:

Does the type of plant change the effect of sunlight?

The unexpected result gave me a new question to investigate.

Without that surprising result, I might never have thought of the new question.

Don't Change the Data

One important rule is that scientists should not change their data just because it doesn't match their prediction.

If the experiment produces an unexpected result, the correct response is to investigate it.

Changing the data would make the experiment less useful because the results would no longer represent what actually happened.

The goal of an experiment isn't to prove that your prediction was correct.

The goal is to learn from the evidence.

Science Doesn't Always Go in a Straight Line

I used to think science was basically:

Question → experiment → correct answer.

But it's often much messier than that.

Scientists can get unexpected results.

They can discover mistakes.

They can change their ideas.

They can run the experiment again.

They can ask new questions.

They can even discover something they weren't originally looking for.

That makes science more interesting because you don't always know exactly where an investigation will lead.

Reflection

Unexpected results used to sound like something I should try to avoid.

Now I think they can be one of the most interesting parts of an experiment.

An unexpected result might reveal a mistake, show that something needs to be tested again, or lead to a completely new question.

The important thing is not to ignore it.

When the results surprise you, don't just ask "What went wrong?" Ask "What can this teach me?"