Measurements, Models & Evidence · Grade 9 Science
Chapter: Investigating EarthMeasurements, Models & Evidence
Learning goals
- I can distinguish a measurement from an inference and report appropriate units.
- I can evaluate a model using evidence, uncertainty, and its limits.
Readings for this unit
Before this unit
Compute a mean, read axes and distinguish observation from inference.
Plan three teaching sessions: core explanation and Parts 1–2; Parts 3–4 with diagrams and reasoning checks; then the whole-unit application, worked examples, independent practice, and exit check. Read one part at a time, explain it without notes, and revisit any gap. Allow about 40–50 minutes per session; adjust after checking understanding. Trace and explain the labeled figures in each part. In the third session, complete the illustrated case and changed independent application; retain a corrected explanation rather than only an answer.
Start with something to explain
How can repeatable measurements still mislead us?
Illustrative evidence case. A balance gives 9.6 g in three trials for a reference labeled 10.0 g; a cloudy stream has different upstream and downstream concentrations.
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.
How can repeatable measurements still mislead us?
Your first model
Sketch observation, proposed explanation and a test that could distinguish alternatives.
Build the language from the idea
- observation
- Information gathered through senses or instruments.
- inference
- An interpretation based on evidence and reasoning.
- uncertainty
- A description of limits on how precisely or confidently a value is known.
Earth scientists cannot directly observe every place or time they study. A thermometer reading is a measurement; an explanation of what caused a temperature change is an inference. Good explanations connect evidence to a mechanism and consider alternatives. A graph or computer simulation is a model, not the Earth itself.
Report a value with units and a precision supported by the instrument. Repeating observations helps reveal variability but does not automatically remove a consistent instrument error. Compare methods, calibration, sample size, and the range of conditions. A mean summarizes measurements; the spread also matters.
Choose a system boundary and decide which matter and energy transfers to include. A model may work for a regional weather pattern yet fail at one street corner. Before using a diagram, table, or reference chart, read its title, axes, units, key, and conditions. State what the evidence supports and what remains uncertain.
A measurement always has a method and uncertainty. A ruler marked every millimeter cannot support a length reported to a millionth of a millimeter. Repeated trials reveal spread, while a calibration check can reveal a common bias that repetition does not remove.
Graphs make comparisons easier when both axes have quantities, units and scales. A straight trend supports a useful relationship over the tested range; extending it far beyond that range is an assumption. An unusual point deserves investigation rather than automatic deletion.
Separate what was measured from the explanation. Several mechanisms may fit a small dataset. State a claim, cite the values that support it, connect them with a mechanism and identify a test that could distinguish an alternative explanation.
Build your understanding
Plan three teaching sessions: core explanation and Parts 1–2; Parts 3–4 with diagrams and reasoning checks; then the whole-unit application, worked examples, independent practice, and exit check. Read one part at a time, explain it without notes, and revisit any gap. Allow about 40–50 minutes per session; adjust after checking understanding.
Part 1 · From observation to a testable explanation
Suspended material consists of particles carried within water rather than dissolved as separate molecules or ions. A concentration of 28 mg/L means 28 milligrams of collected material per litre in the stated measurement. The downstream sample contains more than the upstream sample, but that observation does not identify which land area supplied it. Rain-driven runoff, an eroding bank and a disturbed channel can produce similar appearances. Compare samples taken at the same time, record rainfall and inspect possible pathways before choosing a cause.
Read the visual model
Follow the stream direction and compare the two measurements. The rain arrow proposes one pathway; it is not proof that every downstream particle came from that source.
Which statement is directly supported: downstream concentration is higher, or a particular construction site caused it?
Visual reasoning — after your attempt
Only the higher downstream concentration is directly supported by the supplied measurements. Identifying a source needs upstream/downstream controls, timing and material or pathway evidence.
An observation records an event or property; an inference proposes what it means. State both separately so another person can check the link. A scientific question identifies something measurable, a comparison, and conditions. A claim about a cause needs evidence that helps distinguish it from other possible causes.
If a stream becomes cloudier after rain, the observation is increased cloudiness. Runoff carrying sediment is one possible explanation. A changed instrument, upstream construction, or disturbed streambed could also matter; record what each explanation predicts.
Pause and explain
Why does a cloudy stream observation not establish its cause?
Check your reasoning after answering
Several processes can produce cloudiness; additional measurements or comparisons must distinguish their predictions.
Part 2 · Measurements, precision and repeated trials
Precision concerns how close repeated measurements are to one another; accuracy concerns agreement with an appropriate reference or accepted value. The balance repeatedly reads 9.6 g for a reference mass of 10.0 g. The readings are repeatable but biased downward by 0.4 g. Calibration checks the instrument against a known reference and can reveal this systematic error. Random variation and a calibration offset are different problems: averaging repeated readings can reduce some random variation without removing a constant offset.
Read the visual model
The three dots lie together, while the dashed reference lies above them. Compare both the spread among dots and their distance from the reference.
Would averaging these three readings give the correct 10.0 g mass?
Visual reasoning — after your attempt
No. Their average remains 9.6 g. A reference check reveals the offset; additional identical readings do not remove it.
A measurement has a value, unit, method, and uncertainty. Resolution is the smallest change an instrument can show. Repeated measurements reveal variability, but a consistently miscalibrated instrument can repeat the wrong value. Match reported precision to the method instead of treating many decimal places as greater accuracy.
Read a ruler to the precision its markings support. If three readings cluster but disagree with a checked reference, investigate calibration rather than simply average away the difference. Record units before calculating.
Pause and explain
Can three identical readings prove an instrument is accurate?
Check your reasoning after answering
No. Repetition shows consistency; comparison with a suitable reference helps test accuracy.
Part 3 · Build and evaluate a model
A mathematical runoff model might assume that a fixed fraction of rain reaches the collection cup. Its prediction is conditional on that fraction, collection time and system boundary. In the supplied comparison, the 80% model predicts 80 mL from 100 mL of rain, while only 40 mL is collected. The other water may enter pores, remain on the surface, evaporate or escape collection. Test those stores and losses instead of simply declaring the missing water destroyed. A revised model should explain the mismatch and predict a new trial.
Read the visual model
Read model and observation at the same horizontal rain value. The vertical gap is a difference in collected runoff, not a difference in the original rain supplied.
At 150 mL supplied, what is the runoff mismatch and what measurement could test an explanation?
Visual reasoning — after your attempt
The model gives 120 mL and the comparison gives 60 mL, a 60 mL mismatch. Measure infiltration, surface storage and collection losses under matched conditions to test the assumed runoff fraction.
A model selects useful features of a system: a map, mathematical relationship, physical arrangement, or simulation. Identify what its symbols represent, the assumptions, and what it omits. Compare its prediction with observations under the same conditions; a mismatch can identify an inadequate assumption, measurement problem, or missing process.
A runoff model that assumes all rainfall leaves immediately cannot describe storage and infiltration. Adding those pathways changes the predicted discharge. More detailed models still require evidence, not acceptance because they look realistic.
Pause and explain
What would you check when predicted runoff exceeds measured runoff?
Check your reasoning after answering
Check time interval, rainfall/discharge measurements, storage and infiltration assumptions, and the modeled boundary.
Part 4 · Turn evidence into an argument
A controlled comparison keeps important conditions comparable while changing the factor being tested. To investigate airflow and evaporation, use equal exposed water area, initial temperature, water amount and measurement time. Weigh the same container-plus-water system at the start and later times. If the fan setting is the planned difference, unequal dish area would provide another possible explanation for unequal mass loss. Repeated matched trials help show whether the difference is stable; a causal argument connects the comparison to the particle process of evaporation.
Read the visual model
Compare equal time intervals, then check whether dish area and temperature were held comparable. The graph alone cannot verify the control conditions.
How much greater is loss in A after three hours, and why must dish area be controlled?
Visual reasoning — after your attempt
A loses 6 g and B 3 g, a 3 g difference. Larger exposed area could independently increase evaporation, so unequal area would confound the airflow comparison.
A scientific explanation connects a claim to relevant observations through a physical process. Show why the evidence supports the claim and what remains uncertain. A comparison is strongest when important conditions are controlled or measured. Distinguish lack of evidence for a claim from evidence that rules it out.
A graph should identify variables and units; its pattern needs an explanation. If hotter samples evaporate faster, repeat the comparison with equal water area and airflow before isolating the role of temperature.
Pause and explain
Why are equal exposed area and airflow important in an evaporation test?
Check your reasoning after answering
Both can affect evaporation, so uncontrolled differences weaken an attribution to temperature.
Apply the whole unit · Repeated temperatures and a hidden bias
Illustrative teaching data: a checked reference reads 20.0 °C. A sensor reads 22.1, 22.0 and 22.2 °C in the same stable water bath.
Your task
Calculate the sensor mean and mean difference from the reference. Explain precision versus accuracy, propose a calibration check, and state what these readings cannot establish.
What your response should show
- Use the supplied evidence accurately and label units or locations.
- Explain a physical mechanism or evidence-to-claim relationship.
- Show calculations or a clearly labeled model when the task calls for one.
- Identify a meaningful assumption, uncertainty, or competing explanation.
Response guidance — after your attempt
Mean = 22.1 °C; difference = +2.1 °C. The readings are consistent but disagree with the reference. Check both devices and conditions; the result suggests a systematic offset but does not prove the offset is identical at every temperature.
Scientific references and adapted source material
Parent / teacher teaching notes
Ask the learner to distinguish observation, inference and recommendation aloud. Reteach by changing one condition and asking which conclusion remains justified. Do not reward additional decimals without instrument support. Ask the learner to explain each visual label and causal step before calculating. Preserve a first model, attempt the changed independent case, then use feedback to revise. Diagrams state whether values are illustrative or observed. Selected OpenSciEd excerpts are middle-school source material used for supported foundations; the independent high-school explanations and supplements are not the complete OpenSciEd teacher editions or an endorsement.
The key idea
A scientific explanation links measured evidence to a model while making assumptions and limitations explicit.
Worked examples
Example 1
- Mean = (12 + 14 + 13) ÷ 3 = 13 mm.
- The observed range is 12–14 mm.
- State the method and whether the values are repeated measurements or different sites.
Example 2 · Worked application
- The spread is only 0.2 cm, suggesting good repeatability.
- Accuracy requires comparison with a reliable reference, not just agreement among readings.
A miscalibrated instrument can give consistent wrong readings.
Example 3
- Mean=(18+20+22)/3=20 mm.
- Observed range is 18–22 mm; only three measurements and possible instrument bias limit the result.
Example 4 · Find and fix the mistake
- Test the claim against the stated conditions: A miscalibrated instrument can give consistent wrong readings.
- Precision concerns spread; accuracy concerns closeness to a reference. A systematic offset can affect every reading.
A miscalibrated instrument can give consistent wrong readings.
Example 5 · Reason from the visual evidence
- Read model and observation at the same horizontal rain value. The vertical gap is a difference in collected runoff, not a difference in the original rain supplied.
- The model gives 120 mL and the comparison gives 60 mL, a 60 mL mismatch. Measure infiltration, surface storage and collection losses under matched conditions to test the assumed runoff fraction.
Both lines are deliberately supplied model/test records rather than field observations. An 80% runoff assumption is compared with a 40% outcome over the same rainfall range. Connecting the points expresses those specified constant fractions; an actual landscape may respond nonlinearly.
Avoid these mistakes
- A precise-looking average can still be inaccurate if the instrument is biased.
Check your understanding
Three rain measurements are 18, 20 and 22 mm. Report their mean and range, and describe a limitation.
Discuss the answer
Mean 20 mm; range 18–22 mm; these values do not establish long-term rainfall.
Why does averaging repeated measurements not always remove instrument error?
Discuss the answer
A consistent calibration offset remains in every repeat and their average. Compare suitable reference standards and distinguish systematic bias from random measurement variation.
Return to the phenomenon
Return to the initial case. How can repeatable measurements still mislead us? Revise your drawing or explanation using two named ideas and one stated limitation.
Compare explanations after your attempt
Agreement among repeats does not remove a calibration bias, and a stream-concentration difference does not identify a cause by itself. Test references and controlled comparisons, then connect them to a physical process.
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 the tasks after studying the visual parts. Keep the first model and explain a revision before consulting the guidance.
Activity 1 · Explain the opening case
Use this unit’s opening record, Parts 1–4 and your notebook model. No external equipment or text is required for this supplied-data route.
Your task
Return to the initial case. How can repeatable measurements still mislead us? Revise your drawing or explanation using two named ideas and one stated limitation.
Success criteria
- Use two stated observations or conditions.
- Connect the mechanism to a labeled diagram or calculation.
- Identify what your original model missed.
- State a useful further test and a limitation.
Hint
Name the quantities and interactions before assigning a cause.
Response guidance — discuss after your attempt
Agreement among repeats does not remove a calibration bias, and a stream-concentration difference does not identify a cause by itself. Test references and controlled comparisons, then connect them to a physical process.
Activity 2 · Transfer to a changed case
Use the figures, captions and vocabulary in this unit; cover the worked answer.
Your task
A balance repeatedly reads 19.7 g for a 20.0 g reference. A student reports their average as accurate because the readings agree. Explain the error and propose a correction check.
Success criteria
- Use the stated measurements or conditions, with correct units.
- Connect the result to the physical mechanism or evidence relationship.
- Identify the assumption, model limit or further evidence requested.
Hint
Separate measured values, calculated results and model assumptions.
Response guidance — discuss after your attempt
The readings are precise but 0.3 g below the reference. Their average is still 19.7 g. Check calibration with suitable reference masses over the working range and repeat the measurement; do not assume repeatability proves accuracy.
Words to know
- observation
- Information gathered through senses or instruments.
- inference
- An interpretation based on evidence and reasoning.
- uncertainty
- A description of limits on how precisely or confidently a value is known.
In the real world
Exit check
Name one observation, one inference, and one limitation for a cooling-water investigation.
Check your explanation after answering
Observation: temperature fell from 25°C to 22°C in ten minutes. Inference: energy transferred to cooler surroundings. Limitation: the thermometer has finite resolution or the room temperature was not controlled.
Explain your reasoning without notes. If a key idea is missing, revisit the lesson and try again.