Designing a Biology Investigation — Print · Grade 10 Science
Chapter: Investigating the Chemistry of LifeDesigning a Biology Investigation
Learning goals
- I can identify an independent variable, an outcome, and useful controls.
- I can interpret repeated measurements without claiming more than the evidence supports.
Readings for this unit
Before this unit
Compute a mean and separate measured evidence from inference.
Use three flexible study sessions: core explanation and Parts 1–2; Parts 3–4 with their reasoning checks and models; then the whole-unit application, worked examples, independent practice, and exit check. Allow about 50–60 minutes per session, including diagrams, data work, and notebook explanations. Attempt each check before opening its guidance. Schedule chapter investigations, review, and exams separately; add time where prerequisites are uncertain. Trace the explained visual parts and write a first model, then attempt the illustrated and changed independent cases. In the third session explain a correction and revise the opening account using evidence.
Start with something to explain
Does a taller plant show that its nutrient treatment caused more growth?
Illustrative evidence case. Groups can differ in starting size and environmental conditions; repeated readings can also be mistaken for independent plants.
Notice and wonder
Record two observations using the stated quantities or relationships. Keep observations separate from possible causes. Write one question that a new measurement could help answer.
Does a taller plant show that its nutrient treatment caused more growth?
Your first model
Draw treatment units, a control and the measured change, then identify a confound.
Build the language from the idea
- independent variable
- The factor deliberately changed in a controlled test.
- replicate
- An independent repetition of an experimental condition.
- confounding variable
- Another changing factor that could explain the measured outcome.
A biological investigation starts with a question that can be answered using evidence. To test how temperature affects enzyme activity, change temperature and measure an agreed outcome, such as product formed per minute. Keep enzyme amount, substrate amount, pH, and measurement time comparable. A comparison condition helps interpret the result; it is not the same thing as every controlled variable.
Independent replicates reveal variation. Repeating a reading of one sample tests measurement repeatability; using independently prepared samples tests more of the biological and preparation variation. State which kind you used. A mean summarizes a group, but the spread and sample size matter too. A small difference between means can be unconvincing when variation is large.
An association between two measurements does not by itself establish cause. Check whether the proposed mechanism fits, whether other factors changed, and whether the result repeats. A model simplifies a system and must be tested against observations. Explain uncertainty, alternative explanations, and a next test rather than calling a single result proof.
Turn a broad curiosity into a question specifying organism, conditions and outcome. In a controlled experiment, deliberately vary the independent variable and measure the dependent variable. Record controls that help isolate the proposed cause and distinguish them from a comparison group.
Independent replicates capture biological and preparation variation. Several readings from one leaf are not equivalent to testing several independently grown plants. Report both sample number and spread, and explain whether the samples reasonably represent the population of interest.
A graph can reveal a pattern without proving cause. Random assignment, repeated trials and control of confounders strengthen causal interpretation. When experimentation is unavailable, observational evidence still matters, but the conclusion should match what the design can establish.
Build your understanding
Use three flexible study sessions: core explanation and Parts 1–2; Parts 3–4 with their reasoning checks and models; then the whole-unit application, worked examples, independent practice, and exit check. Allow about 50–60 minutes per session, including diagrams, data work, and notebook explanations. Attempt each check before opening its guidance. Schedule chapter investigations, review, and exams separately; add time where prerequisites are uncertain.
Part 1 · Turn a question into a testable claim
A hypothesis proposes an explanation that leads to a testable prediction. An operational definition says exactly how a quantity will be measured. Here growth means change in seedling height over three days, not the largest final plant. Starting height, species and the observation interval matter. Nutrient treatment is the explanatory variable; height change is the response. A predicted difference is not yet evidence that a nutrient causes growth, because light, water or starting condition could also differ.
Read the visual model
Compare changes from the common starting value and retain the time unit.
How much growth does each supplied group show, and what would weaken a nutrient-only explanation?
Visual reasoning — after your attempt
The changes are 6 and 15 cm; their difference is 9 cm. Unequal water, light, species or initial condition would confound the causal comparison.
A biological question becomes testable when you can name one thing you will change, one thing you will measure, and the things you will hold constant. The variable you change is the independent variable; the measured outcome is the dependent variable; everything deliberately held steady is a controlled variable. A hypothesis states the expected relationship between the first two and gives a reason drawn from biology, not a guess about which result would be more interesting.
"Does light affect plants?" cannot be tested as written, because neither the change nor the measurement is specified. "Do radish seedlings grown at 12 hours of light per day produce greater mean dry mass after 14 days than seedlings grown at 6 hours?" names the organism, the independent variable with its levels, the measured outcome, and the duration. A reader can repeat it.
Pause and explain
Why does a hypothesis need a biological reason and not only a prediction?
Check your reasoning after answering
A prediction alone cannot be evaluated when the result is unexpected. A stated reason — more light allows more photosynthesis, so more mass accumulates — tells you which part of your thinking the data contradicted.
Part 2 · Build controls and replication into the design
A biological replicate is an independently sampled or treated biological unit. Measuring the same plant four times supplies repeated readings of one plant, not four independent plants. Random assignment can distribute known and unknown starting differences across treatments; it cannot guarantee perfectly identical groups. A control group provides a suitable reference under comparable handling. If all treated plants share one tray and all controls another, a tray effect remains mixed with treatment, so independence and the treatment level must be stated.
Read the visual model
Count distinct plants and identify whether treatment was assigned by plant or tray.
Why do twelve readings of one plant not establish twelve independent treatment responses?
Visual reasoning — after your attempt
They share the same plant history and treatment unit. They can improve a time record but cannot substitute for independently treated biological replicates.
A control group experiences every condition except the one being tested, so it shows what happens without the treatment. Replication means independent repeats: several seedlings, several dishes, several trials. Measuring the same single plant five times gives you five readings of one individual, which describes your measuring, not the biological variation among plants. That distinction is the one most home investigations get wrong.
Sample size matters because living things vary. With two plants per group, one unusually vigorous seedling can produce the whole difference you observe. With eight per group, a single outlier moves the mean much less. State the number of independent units per group before you collect data, and keep every one of them in the record, including the ones that died.
Pause and explain
You measure one leaf's length three times and get 6.1, 6.0, and 6.2 cm. What have you established?
Check your reasoning after answering
The precision of your measurement of that leaf, about ±0.1 cm. You have established nothing about how leaf length varies among plants, which needs separate leaves on separate plants.
Part 3 · Summarize data honestly
A mean is the sum of measured values divided by their number. The range describes the spread from minimum to maximum; it is not an uncertainty interval for a population mean. These four plants have mean 25 mm and range 10–40 mm. A bar showing only the mean would hide the variation. Report sample size, units and individual results or a useful distribution plot. Do not discard a low result solely because it weakens the preferred hypothesis; exclusions require a stated reason such as a documented measurement failure.
Read the visual model
Read all four measurements before averaging; none is an automatic error.
Would replacing these four records with “25 mm for every plant” be accurate?
Visual reasoning — after your attempt
No. Twenty-five is the mean, while the individual plants range from 10 to 40 mm. Retain variation and the sampling conditions.
Report a measure of centre and a measure of spread together. The mean summarizes a group; the range or standard deviation shows how much the individuals disagreed. A difference between two means is only interesting when it is large compared with the spread inside each group. Two groups with means of 12 cm and 13 cm, each ranging from 8 to 17 cm, overlap so heavily that the one-centimetre difference carries little weight.
Choose the graph that fits the data. A line graph implies a continuous relationship, so use it when the independent variable is a quantity such as temperature or time. A bar graph suits discrete categories. Label both axes with units, show the levels you actually tested, and do not extend a line beyond the range you measured.
Pause and explain
Group A has a mean of 20 units with a range of 19–21. Group B has a mean of 22 units with a range of 10–34. Which difference is better supported?
Check your reasoning after answering
Neither group alone settles it, but B's enormous spread means its mean is poorly determined; the two-unit difference is small next to B's 24-unit range. Collect more replicates in B before claiming a treatment effect.
Part 4 · Write the argument, including its limits
A claim answers the investigation question. Evidence consists of relevant observations; reasoning explains how those observations constrain a mechanism or alternative. Here treatment and comparison responses are similar across trials, so a large consistent benefit is not supported by this supplied record. That does not prove the treatment can never matter under other conditions. Separate an observed difference, a statistical inference and a proposed cause. A useful follow-up changes one relevant factor, preserves comparable conditions and specifies the predicted outcome before results are collected.
Read the visual model
Compare paired results rather than selecting the single greatest treatment value.
What claim is more defensible than “the treatment always produces a large benefit”?
Visual reasoning — after your attempt
The supplied matched trials do not show a large consistent benefit. A broader claim requires replicated comparisons across defined conditions and uncertainty analysis.
A scientific argument has three parts: a claim that answers the question, evidence drawn from your own recorded data, and reasoning that connects the evidence to the claim through biology. Numbers alone are not an argument, and a restatement of the hypothesis is not a conclusion. Quote the means and spreads you measured, then say which biological mechanism accounts for them.
Every investigation has limits, and naming them strengthens the report. Say what you could not control, how many units you had, how precisely you could measure, and what a reasonable alternative explanation would be. Then name the next measurement that would distinguish your explanation from that alternative. A report that claims more than its design can support is weaker than one that states a narrow, supported result.
Pause and explain
Why is "more research is needed" a weak statement of limitations?
Check your reasoning after answering
It identifies nothing. A useful limitation names the specific uncontrolled factor, the specific alternative explanation, and the specific additional measurement that would separate them.
Apply the whole unit · Repair a flawed growth study
Illustrative record from a home investigation. Group 1: one seedling given fertilizer, on a south windowsill, final height 19 cm. Group 2: one seedling given plain water, in a north corner, final height 11 cm. Both were watered "when the soil looked dry." The student concluded that fertilizer increases growth by 73 percent.
Your task
Identify every variable that differs between the groups, explain why the 73 percent figure cannot be assigned to fertilizer, and rewrite the design so that the same question becomes testable. State your sample size, your controlled variables, and the two numbers you would report for each group.
What your response should show
- Choose the appropriate biological model or relationship and state the conditions it assumes.
- Show the counts, rates, ratios, or intermediate reasoning the question requires.
- Use the supplied evidence to justify the conclusion and separate observation from inference.
- Identify a meaningful limitation, uncertainty, or next measurement.
Response guidance — after your attempt
List the differences: fertilizer, light exposure, and watering schedule all vary at once, so three candidate causes are confounded.
Note the sample size: one plant per group means zero replication, so no variation among individuals is measurable and any single vigorous plant determines the result.
Compute what the percentage rests on: (19 − 11) / 11 × 100 ≈ 73 percent, derived from exactly two measurements, so it carries no information about spread.
Rewrite: eight seedlings per group, all on the same windowsill, same pot size and soil, a measured 50 mL of water on a fixed schedule, with fertilizer concentration as the only difference; report each group's mean height and range after 14 days.
Fertilizer, light, and watering differ together, so the taller plant cannot be attributed to fertilizer; light is the most likely alternative explanation because it drives photosynthesis directly. With one plant per group there is no replication, so the 73 percent difference describes two individuals rather than two treatments. A supported design holds light, water volume, soil, and pot size constant, varies only fertilizer concentration, uses at least eight independent seedlings per group, and reports the mean and range of final height for each. Even then the conclusion applies to radish seedlings under these conditions, not to plants in general.
Scientific references and adapted source figures
Parent / teacher teaching notes
This unit sets the standard the whole course is marked against. Insist on the distinction between repeated measurements of one organism and independent replicates, because it recurs in every later investigation. When your child reports a difference between two means, ask for the spread in each group before discussing the mechanism. Accept a narrow supported conclusion over a sweeping one. Ask for an explanation of each label, arrow and axis before calculation. Keep first and revised models; attempt the changed case before consulting guidance. Distinguish invented instructional records from named historical sources. Source chromosome symbols are allele-category models, not DNA sequences. These independently authored high-school supplements are not the full OpenSciEd teacher editions or an endorsement.
The key idea
A strong claim names the tested conditions, cites data, explains a mechanism, and states limits.
Worked examples
Example 1
- Mean = (8 + 10 + 9) ÷ 3 = 9 units/min.
- Report the observed range, 8–10 units/min, and number of replicates.
- Do not infer an exact universal rate from this small group.
Example 2 · Worked application
- Light and water both vary, so either can explain growth differences.
- Hold water, plant type, soil and starting size comparable while varying light.
- Use replicated plants and a stated growth measure.
Replication helps assess variation but does not by itself remove confounding.
Example 3
- The ten readings describe repeatability for one plant.
- They do not capture variation among plants or provide an untreated comparison.
Example 4 · Find and fix the mistake
- Test the claim against the stated conditions: Replication helps assess variation but does not by itself remove confounding.
- More samples do not fix systematic differences in conditions. Control the competing variable.
Replication helps assess variation but does not by itself remove confounding.
Example 5 · Reason from the visual evidence
- A mean is the sum of measured values divided by their number. The range describes the spread from minimum to maximum; it is not an uncertainty interval for a population mean. These four plants have mean 25 mm and range 10–40 mm. A bar showing only the mean would hide the variation. Report sample size, units and individual results or a useful distribution plot. Do not discard a low result solely because it weakens the preferred hypothesis; exclusions require a stated reason such as a documented measurement failure.
- No. Twenty-five is the mean, while the individual plants range from 10 to 40 mm. Retain variation and the sampling conditions.
Original illustrative replicate outcomes. They are not evidence of a universal biological growth rate or a significance test.
Avoid these mistakes
- Repeated readings of one sample are not independent biological replicates.
Check your understanding
One plant is measured ten times after treatment. Does this establish the treatment's effect on plants generally?
Discuss the answer
No; use independent plants, appropriate comparisons and controlled conditions to test a broader claim.
Why can random assignment strengthen an experimental comparison?
Discuss the answer
Random assignment helps distribute starting differences across treatment groups, reducing confounding on average; it does not guarantee identical groups or replace independent replication.
Return to the phenomenon
Return to the initial case. Does a taller plant show that its nutrient treatment caused more growth? Revise your drawing or explanation using two named ideas and one stated limitation.
Compare explanations after your attempt
A defensible comparison defines growth, treatment assignment, independent replication and comparable conditions; a mean does not replace variation or demonstrate a unique cause.
What should we investigate next?
Which comparison would test a remaining uncertainty? State what you would change, keep comparable and measure.
Investigate, explain and revise
Complete these tasks after the teaching parts. Preserve your first model and a supported revision; use guidance after a real attempt.
Activity 1 · Return to the opening case
Use the opening question, defined terms, Parts 1–4 and your notebook. The supplied-data route needs no equipment or external account.
Your task
Return to the initial case. Does a taller plant show that its nutrient treatment caused more growth? Revise your drawing or explanation using two named ideas and one stated limitation.
Success criteria
- Use relevant stated observations.
- Explain the mechanism with a labeled model.
- Distinguish quantities and assumptions.
- Describe a revision and a useful next test.
Hint
Follow objects, materials and information separately. Check what each measured quantity means.
Response guidance — discuss after your attempt
A defensible comparison defines growth, treatment assignment, independent replication and comparable conditions; a mean does not replace variation or demonstrate a unique cause.
Activity 2 · Transfer to a changed case
Cover the worked guidance and use the unit explanations/figures to solve a changed problem.
Your task
Two trays receive different nutrient treatments. Each tray contains one plant measured six times. Explain the true replication and redesign the comparison.
Success criteria
- Use the stated measurements or conditions, with correct units.
- Connect the result to the biological mechanism or evidence relationship.
- Identify the assumption, model limit or further evidence requested.
Hint
Name the stated conditions, select the relevant mechanism and show intermediate reasoning.
Response guidance — discuss after your attempt
There is only one treated biological/treatment unit per condition; six readings are repeated observations. Use several independently assigned plants or trays per condition, keep light/water/species comparable and preserve all time records.
Words to know
- independent variable
- The factor deliberately changed in a controlled test.
- replicate
- An independent repetition of an experimental condition.
- confounding variable
- Another changing factor that could explain the measured outcome.
In the real world
Exit check
Describe one fair comparison for investigating enzymes.
Check your explanation after answering
Compare equal enzyme and substrate amounts at different temperatures while holding pH and time constant; measure product per time using independent replicates.
Explain your reasoning without notes. If a key idea is missing, revisit the lesson and try again.
Problem 1
Answer and explanation
6 units/min
18 ÷ 3 = 6.
Problem 2
- Enzyme amount
- The measured reaction rate
- The graph height
- The conclusion
Answer and explanation
Enzyme amount
Changing enzyme amount would introduce another potential cause.
Problem 3
Answer and explanation
Repeated readings emphasize measurement variability; independent samples also reflect preparation and biological variation.
Problem 4
Answer and explanation
No; use independent plants, appropriate comparisons and controlled conditions to test a broader claim.
The ten readings describe repeatability for one plant.
They do not capture variation among plants or provide an untreated comparison.
Response criteria
- Use the stated conditions.
- Show the evidence or calculation.
- Explain the conclusion and any stated limitation.
Problem 5
Hints
Replication helps assess variation but does not by itself remove confounding.
Answer and explanation
Use several comparable plants per treatment, assigned fairly, with other conditions held constant.
Problem 6
Hints
State what the supplied observations or conditions establish, and separate that from any assumption.
Answer and explanation
An operational measure such as height change or dry mass allows repeatable comparisons; appearance alone is subjective.
Problem 7
Answer and explanation
No. Twenty-five is the mean, while the individual plants range from 10 to 40 mm. Retain variation and the sampling conditions.
A mean is the sum of measured values divided by their number. The range describes the spread from minimum to maximum; it is not an uncertainty interval for a population mean. These four plants have mean 25 mm and range 10–40 mm. A bar showing only the mean would hide the variation. Report sample size, units and individual results or a useful distribution plot. Do not discard a low result solely because it weakens the preferred hypothesis; exclusions require a stated reason such as a documented measurement failure.
No. Twenty-five is the mean, while the individual plants range from 10 to 40 mm. Retain variation and the sampling conditions.
Response criteria
- Read the structures, axes, directions and units correctly.
- Connect the evidence with the biological mechanism.
- Respect stated model assumptions and distinguish an observation from a further inference.
Problem 8
Answer and explanation
There is only one treated biological/treatment unit per condition; six readings are repeated observations. Use several independently assigned plants or trays per condition, keep light/water/species comparable and preserve all time records.
There is only one treated biological/treatment unit per condition; six readings are repeated observations. Use several independently assigned plants or trays per condition, keep light/water/species comparable and preserve all time records.
Response criteria
- Use the stated measurements or conditions, with correct units.
- Connect the result to the biological mechanism or evidence relationship.
- Identify the assumption, model limit or further evidence requested.