Reading an MCAT experiment without getting lost in the details

The experiment underneath the terminology

A science passage can feel difficult before you reach the first question. There are unfamiliar names, a method you have never used and a table that seems to add even more information. Before you reread the whole thing, check whether you can explain what the researchers actually did.

I would start with a few concrete questions: What did they change or observe? What did they measure? Which groups or conditions did they compare? How, exactly, did they define the outcome?

You still need the relevant science. This gives that science a structure. When a question sends you back to a graph or method, you have a better idea of what you are looking for.

A short original example

Researchers randomly assign seedlings of the same species to receive either four or eight hours of light each day. They keep soil, water and temperature the same. After two weeks, they compare the average increase in stem length, measured in centimeters.

Pause before reading on. Explain the study in your own words without looking at the paragraph. Then look back and check what you left out.

What the study actually measures

What changed? The assigned number of hours of light per day. That is the independent variable in this experiment.

What was measured? The increase in stem length over two weeks. That is the dependent variable.

What is the comparison? Average growth in the eight-hour group versus average growth in the four-hour group.

How is growth defined? Final stem length minus starting stem length, in centimeters. The outcome is a change over a stated period.

Here are made-up results for practice. The numbers illustrate the reasoning, rather than report a real plant study.

Light per day: 4 hours
Mean starting length: 10 cm
Mean final length: 14 cm
Mean increase: 4 cm

Light per day: 8 hours
Mean starting length: 10 cm
Mean final length: 17 cm
Mean increase: 7 cm

The average increase is 3 cm greater in the eight-hour group: 7 minus 4. The described design tests the effect of light duration on this outcome under these conditions. These averages alone do not tell us whether the difference is statistically significant. We would need additional information about the data and analysis.

Final height and growth answer different questions

Now consider two individual plants in a separate example. Plant A starts at 10 cm and ends at 14 cm. Plant B starts at 15 cm and ends at 17 cm. B ends taller, but A grows more: 4 cm compared with 2 cm.

If you answer a growth question using only final height, you can choose the wrong comparison even though you read every number correctly. Check the outcome definition before calculating. Apply the same care when a passage measures total product instead of reaction rate, or the percentage of people meeting a threshold instead of the size of each person’s change.

What changes in an observational study

Suppose researchers instead visit existing gardens, record daily light exposure and measure plant growth. They do not assign the amount of light. Gardens with more light might also receive different amounts of water or contain different soil.

A relationship between light and growth would not, by itself, isolate the effect of light. Describe what the researchers observed, and consider what else differs between the groups. Do not add random assignment or controlled conditions that the passage never gives you.

A quick check

In a new experiment, one group receives fertilizer and 100 mL of water daily. A second group receives no fertilizer and 50 mL of water daily. The first group grows more. Can you attribute the difference specifically to fertilizer?

No. Fertilizer and water both differ between the groups. The comparison cannot separate their effects. A useful next comparison would keep water the same while varying fertilizer, with other relevant conditions controlled and appropriate assignment to groups.

Notice what the repair requires. You do not need another definition of fertilizer. You need to check which conditions differ before making a causal claim.

A useful next step in practice

On your next research passage, try one short sentence: “They changed ___, measured ___ and compared ___ with ___.” If nobody manipulated anything, say what they observed instead. Add a brief definition of the measurement when the outcome is easy to confuse.

Use the methods, figure titles, axis labels and footnotes to check your sentence. Then answer the actual question. This is a reading prompt, not a requirement to write a paragraph of notes for every passage.

During review, identify the exact point that went wrong. Did you confuse the outcome with the exposure? Compare final values when the question asked about change? Assume that an observed relationship established a cause? Name that step, and practice correcting it when a new passage gives you the same kind of decision.

For the broader review process, see “How to review a missed question so the next attempt changes”.

Exam context: AAMC includes research design, variable relationships, measurement and interpretation of data in its scientific inquiry and reasoning skills. The examples here are original teaching exercises, not official AAMC questions. Official outline.