Questions, Variables & Fair Investigations — Print · Grade 7 Science
Chapter: Scientific Inquiry & Diversity of LifeQuestions, Variables & Fair Investigations
Learning goals
- I can explain questions, variables & fair investigations.
- I can use evidence, a model, and stated assumptions to support a biological explanation.
Before this unit
Use the supplied vocabulary, distinguish an observation from an explanation, and identify any units or model assumptions. Revisit a connected earlier unit if needed; new terms are explained here.
Use three flexible 40–45-minute sessions: core background and Parts 1–2 with their checks; Parts 3–4 and the worked application; then remaining practice, a labeled model, revision, and the exit check. Checks completed during reading can count toward the practice set. Read one short part at a time. Investigations and chapter review/quiz have separate sessions, and flexible weeks allow more time.
In 1747 a naval surgeon named James Lind had a problem that had killed more British sailors than enemy action: scurvy. Everybody had a theory and most of them were wrong. What Lind did aboard HMS Salisbury was unusual. He took twelve sick sailors, put them in the same quarters, gave them the same food, and then split them into six pairs — each pair getting one of the six remedies people were arguing about. Cider for one pair, vinegar for another, seawater, a sulphuric acid elixir, a spice paste, and for the last pair two oranges and a lemon. Within six days the citrus pair were well enough to nurse the others. That is the whole shape of a fair test: change one thing, hold everything else the same, and see what follows.
Lind's design had the three parts every investigation needs. The thing he deliberately changed is the independent variable — the remedy. The thing he measured to see what happened is the dependent variable — how the sailors' condition changed. And everything he deliberately kept the same across all six pairs — the diet, the quarters, the starting severity — are the controlled variables. If you can name those three for an investigation, you can judge whether its conclusion is worth anything. If you cannot, the investigation is probably not answering the question it claims to.
The reason this matters is that a question like "do plants grow better with more water?" is not yet testable. It does not say what would count as better, how much more water, or what else would be held steady. A testable question names the change, the measurement and the conditions: "Does 10 mL of water a day produce more height gain over seven days in bean seedlings than 5 mL, with light, soil, pot size and temperature held the same?" That version can come out wrong, which is the point of it. And it protects you against a confound — a second factor that changed alongside the one you meant to test, so that when a difference appears you cannot tell which factor produced it. A confound does not make results useless; it makes them unable to separate causes, which is a different and more precise complaint.
Build your understanding
Where this fits
This is where the course starts, and it starts with method rather than content on purpose. Everything in the units that follow — cells, body systems, ecosystems, evolution — rests on somebody having designed a comparison good enough to trust. This unit is about how that is done, and the rest of Chapter 1 adds the measuring, the graphing and the reasoning that go with it.
Use three flexible 40–45-minute sessions: core background and Parts 1–2 with their checks; Parts 3–4 and the worked application; then remaining practice, a labeled model, revision, and the exit check. Checks completed during reading can count toward the practice set. Read one short part at a time. Investigations and chapter review/quiz have separate sessions, and flexible weeks allow more time.
Part 1 · Turning a broad question into a comparison
Start with the vague version and interrogate it. "Does fertiliser help plants?" invites four questions back. Which plants? How much fertiliser, compared with how much? Helps in what measurable way — height, leaf count, mass, time to flower? Over how long? Each answer narrows the question until it names a comparison you could actually set up on a windowsill.
The finished version always has the same three slots. What you will change, and to what values: 0 g against 2 g of the same fertiliser. What you will measure, with units: height in centimetres after fourteen days. What you will hold the same: species, pot size, soil, starting height, water volume, light, temperature. Writing those three lists before you begin is the single most useful habit in this course, because it exposes the problems while they are still cheap to fix.
A good test of whether a question is testable is to ask what result would prove you wrong. "Plants like light" cannot be wrong, so it is not yet a scientific question. "Seedlings grown 20 cm from the window will gain more height in fourteen days than seedlings grown 2 m away, with everything else matched" can be wrong, and that is what makes it worth doing.
Pause and explain
Rewrite "does music help plants grow?" as a testable question.
Check your reasoning after answering
Something like: do bean seedlings exposed to music at 60 decibels for two hours a day gain more height over fourteen days than identical seedlings in silence, with species, pot, soil, water, light and temperature held the same? It names what changes, what is measured and in what units, over what period, and what is held constant — so a result could contradict it.
Part 2 · Spotting a confound before it ruins the result
A confound is a second factor that varies along with the one you meant to test. Suppose Group A gets 10 mL of water a day on a bright windowsill and Group B gets 5 mL on a dim shelf. A grows more. Water and light both differed, so the result is real but it cannot be attributed: you have measured the combined effect of two changes and have no way to divide it between them.
Confounds hide in conditions nobody thought to write down. The bright windowsill is also warmer. The pots at the end of the row dry out faster. The seeds from the bottom of the packet were smaller to begin with. The second week was cloudier than the first. Each of these can produce a difference that looks exactly like the one you were testing for, which is why listing the controlled variables beforehand matters more than it sounds.
There are two defences. Match conditions deliberately, so the only planned difference is the independent variable. And randomise what you cannot match — if you have twenty seedlings and two trays, assign them to trays by drawing lots rather than by picking, so that any hidden difference between the seedlings is spread across both groups instead of landing in one.
Pause and explain
A student finds that plants on the kitchen windowsill grow taller than plants in the hallway, and concludes that light causes growth. What has gone wrong?
Check your reasoning after answering
Light is not the only thing that differs between a kitchen windowsill and a hallway. The windowsill is probably warmer, may be more humid from cooking, and the plants there may be watered more often because they are seen more often. Any of these could produce the difference, so the comparison cannot attribute it to light.
Part 3 · Repeating, and recording what you would rather not
Lind's experiment had two sailors per treatment. That is enough to notice something dramatic and far too few to be sure of anything subtle — two people can differ for a hundred reasons that have nothing to do with the remedy. Repetition is the defence against being fooled by one unusual individual. Grow ten seedlings per group rather than one, and the group's average carries information that a single plant never could.
Repeats also give you something a single measurement cannot: a sense of how much results vary on their own. If your five control plants gain 3, 4, 4, 5 and 4 cm, you know the ordinary spread is about 2 cm, so a treatment group averaging 4.5 cm has told you nothing. If the controls gain 3, 4, 4, 5 and 4 cm and the treatment group gains 11, 12, 10, 12 and 11, the difference is far outside the ordinary spread, and that is what makes it convincing.
Record every result, including the ones that went wrong. A seedling that died, a pot that was knocked over, a reading you think you misread — write them down with a note. Discarding inconvenient results quietly is how an honest person produces a false conclusion, because the discards are rarely random; they tend to be the ones that disagreed.
Pause and explain
Why is a single plant in each group not enough, even if you measure it very carefully?
Check your reasoning after answering
Because individual plants differ for reasons unconnected to the treatment — seed size, germination date, position, a damaged root. With one plant per group you cannot tell a treatment effect from ordinary individual variation. Several plants per group give an average and show how much variation there is anyway, which is what a difference has to exceed to be convincing.
Part 4 · Revising a plan when you cannot control everything
Real investigations are done in kitchens and gardens, not sealed chambers, and some factors cannot be held constant. The honest response is not to abandon the comparison but to narrow what you claim from it. If you cannot control temperature, record it, and say that the conclusion holds for the range you observed. If both trays had to sit in the same window, say so, and note that you have not tested whether position mattered.
Stating a limitation is a strength, not an apology. "Watering at 10 mL daily was associated with greater height gain than 5 mL in this trial of ten seedlings over fourteen days, at temperatures between 18 and 22 degrees, with light and soil matched" is a claim somebody can check and build on. "More water makes plants grow" is not, and it is wrong for most plants past a certain point anyway.
The last revision to consider is the cheapest one: run the comparison again. A result that reappears under slightly different conditions is far stronger than one careful run, because the things you failed to control are unlikely to have fallen the same way twice.
Pause and explain
Your windowsill is the only place with enough light, so both groups must share it. What should you do?
Check your reasoning after answering
Run the comparison there, but control position deliberately: assign pots to places at random and rotate them daily so no group sits permanently in the brightest spot. Then record that light was shared and rotated, and limit the claim to those conditions. Saying what you could not control, and what you did instead, is part of the result.
Apply the whole unit · Evaluate a seedling comparison
Illustrative record, not a measured experiment: Group A receives 10 mL of water daily and sits in bright light; Group B receives 5 mL daily and sits in dim light. After seven days, Group A has gained 4 cm in height and Group B has gained 1 cm.
Your task
Identify the intended variable, the measured outcome and the confound. Propose a fairer comparison, and state what the current record does establish.
What your response should show
- Explain the relevant biological process accurately.
- Use the supplied evidence and show calculation steps with units where needed.
- Connect a labeled model or mechanism with the claim.
- State an assumption, evidence limitation, or useful next observation.
Response guidance — after your attempt
Name what the investigator meant to change — the daily water volume, 10 mL against 5 mL — and what was measured, height gain in centimetres over seven days.
Find the second factor that changed with it: light differed too, bright against dim, so two things varied at once.
State the consequence precisely: the 3 cm difference is real, but it cannot be divided between water and light, so neither can be credited with it.
Propose the fix: hold light identical for both groups and vary only water, with several plants per group, the same species, soil, pot size and starting height.
The intended independent variable is the daily volume of water, 10 mL against 5 mL, and the dependent variable is height gain in centimetres over seven days. Group A gained 4 cm and Group B gained 1 cm, a difference of 3 cm. But light was also different — bright for A, dim for B — and that is a confound: two factors changed together, so the record cannot say whether the extra growth came from the water, the light, or some combination. What it does establish is that the whole package of more water plus more light was associated with more height gain here, which is worth knowing and is not the same as a claim about water. A fairer comparison would place both groups under identical light, hold species, soil, pot size, starting height and temperature the same, vary only the water, and use several plants in each group rather than one, so that ordinary variation between individuals can be seen and allowed for. It would also be worth recording the temperature and rotating the pots, since the brighter position is usually the warmer one.
Scientific references for this unit
Parent / teacher teaching notes
Ask for a because-statement and a labeled relationship, not just a memorized word. Work through one calculation step at a time, permit oral or diagram responses, and use flexible sessions for difficult reading. Compare the learner’s actual evidence with the supplied model; do not reward copying or agreement with a personal preference. No private family, genetic, or medical records are required.
The key idea
Science begins with questions about the natural world that observations can help answer. A question such as whether equal-sized paper samples absorb different amounts of water specifies something we can compare. A preference such as which color is nicest needs a stated judgment criterion; a measurement cannot decide everyone’s preference. Asking a testable question does not require predicting the correct answer in advance.
Worked examples
Example 1 · Find the three variables in a real trial
- The independent variable is the factor deliberately varied between groups: which remedy each pair received.
- The dependent variable is what was measured as an outcome: the change in each sailor's condition, judged by fitness for duty.
- The controlled variables are what was held the same: diet, quarters, and selecting sailors whose scurvy was at a similar stage.
A design can be sound in structure and still too small to be convincing. Judge both.
Example 2 · Decide whether a difference can be attributed
- List what differed between the trays: fertiliser, and position — which means light and probably temperature as well.
- Check whether the difference can be split: with two or more factors varying together, any outcome could come from either, or from both.
- Decide what the result does support: the combination was associated with more growth, and nothing more specific.
Ask what else changed. If the answer is anything at all, the result cannot attribute a cause.
Avoid these mistakes
- Mistake: a question is testable because it is about science. "Does music help plants?" names no measurement and no conditions; a testable version states the change, the units, the duration and what is held the same.
- Mistake: a confound makes the data worthless. The 3 cm difference in the seedling record is real — what it cannot do is say which of the two changed factors produced it.
- Mistake: one plant per group is enough if you measure carefully. Careful measurement of one individual still cannot separate a treatment effect from ordinary variation between individuals.
- Mistake: results that went wrong should be left out. Discards are rarely random — they tend to be the ones that disagreed — so leaving them out quietly is how an honest person reaches a false conclusion.
Check your understanding
Why does defining the outcome before testing help?
Discuss the answer
It keeps the comparison consistent and reduces the temptation to choose whichever observation later supports a favorite explanation.
Two plant groups differ in both soil and water. What claim is too strong?
Discuss the answer
A claim that water alone caused their difference is too strong because soil could also influence growth; separate or control the factors.
Words to know
- testable question
- A question naming what will be changed, what will be measured and in what units, and what is held the same — so that a result could contradict it.
- independent variable
- The one factor the investigator deliberately changes, such as the daily volume of water.
- dependent variable
- The outcome measured to see what the change produced, such as height gain in centimetres.
- controlled variable
- A factor deliberately held the same across all groups so it cannot explain any difference found.
- confound
- A second factor that varied alongside the independent variable, so a difference cannot be attributed to either one.
Extend your scientific explanation
Compare a second supplied example or the linked educational reference. Identify a connection, a model limit, and an additional observation that would test the explanation.
In the real world
Use a clear question, reliable observations, and a biological mechanism to evaluate a claim. State what remains uncertain before applying a classroom model to a different setting.
Exit check
Does repeating an unfair test make it fair?
Check your explanation after answering
No. Repetition can reveal variability while repeating the same confound. The method must change to separate the competing explanations.
Explain your reasoning without notes. If a key idea is missing, revisit the lesson and try again.
Problem 1
Answer and explanation
It keeps the comparison consistent and reduces the temptation to choose whichever observation later supports a favorite explanation.
Problem 2
Answer and explanation
A claim that water alone caused their difference is too strong because soil could also influence growth; separate or control the factors.
Problem 3
Answer and explanation
It may represent real variation or reveal a problem worth investigating. Removing it only because it disagrees with a prediction biases the evidence.
Problem 4
Answer and explanation
No. Repetition can reveal variability while repeating the same confound. The method must change to separate the competing explanations.
Problem 5
Answer and explanation
The record shows a 3 cm difference in height gain, but both water and light differ. It cannot isolate water’s effect. A revised comparison would keep light, species, starting conditions, and times comparable while changing only the defined water treatment; independent seedlings and recorded variation would strengthen the evidence.
Identify water amount as the intended factor.
Measure height change in centimeters over seven days.
Recognize light as an additional changed factor.
Match light and other conditions, then use several seedlings per group.
Response criteria
- Explain the relevant biological process accurately.
- Use the supplied evidence and show calculation steps with units where needed.
- Connect a labeled model or mechanism with the claim.
- State an assumption, evidence limitation, or useful next observation.
Problem 6
Answer and explanation
No. Repetition does not remove the linked light difference. Redesign the comparison so light is comparable, and use independent seedlings with recorded starting sizes.
Response criteria
- Explain the relevant biological process accurately.
- Use the supplied evidence and show calculation steps with units where needed.
- Connect a labeled model or mechanism with the claim.
- State an assumption, evidence limitation, or useful next observation.
Problem 7
- Daily water amount
- Height change
- Container material
- Plant species
Answer and explanation
Height change
Height change is the measured response; daily water is the changed factor.
Problem 8
Answer and explanation
5 cm
Final minus initial: 8−3=5 cm.