Connect a model to a measurement in three links: the model, the prediction it makes, and the data that agrees or disagrees. Write that the data is consistent with the model, not that it proves it.
This lesson belongs to the set on materials and technology through practical evidence. It reuses the layer model from explaining alloy properties.
What are the three links?
- Model: a simple picture of the material, such as layers of atoms.
- Prediction: what the model says should happen when something changes.
- Measurement: what the data show when the change is made.
If the measurement agrees with the prediction, the data is consistent with the model. If it disagrees, the model needs another look.
Worked example 1: added atoms and hardness
The model says added atoms of a different size disturb the layers, so the layers slide less easily. The prediction is that more added atoms make the metal harder.
An original test adds an element to a metal and presses a ball into each sample with the same force.
| Added element (% of sample) | Dent diameter (mm) |
|---|---|
| 0 | 3.0 |
| 2 | 2.4 |
| 5 | 2.0 |
Reading the data. The dent shrinks from 3.0 mm to 2.4 mm to 2.0 mm as the added element rises from 0% to 2% to 5%. A smaller dent means a harder sample.
Writing the link. The measured hardness increases as more atoms are added, which is consistent with the model that added atoms disturb the layers.
Stating the limit. Only three amounts were tested, so the data cannot show that hardness keeps rising at 10% or 20%. The model also does not predict the exact dent sizes.
Worked example 2: chain length and softening
A second model says long polymer molecules tangle more, so more heat is needed to loosen them. The prediction is that longer chains soften at a higher temperature.
| Chain length | Softening temperature |
|---|---|
| Short | 60°C |
| Medium | 90°C |
| Long | 120°C |
The softening temperature rises by 30°C from short to medium, and by another 30°C from medium to long. The data is consistent with the tangling model, although the model does not explain why the steps are equal.
The mistake that costs marks
The common slip is to write “this proves the model is correct”. Three data points can agree with this model and with a different explanation too.
| Wrong | Right |
|---|---|
| “The data proves the atom model” | “The data is consistent with the atom model” |
| “Hardness depends only on added atoms” | “In this test, hardness rose as added atoms rose” |
| “The model explains everything” | “The model predicts the trend but not the exact values” |
The repair is to use “consistent with” or “supports” and to name one limit of the test.
Check yourself
A model says that ceramics are brittle because their atoms are held in a rigid pattern that cannot slide. A test bends strips of Ceramic P, Polymer Q and Metal R until each fails. P snaps at a bend of 2°, Q bends 90° without breaking (the test stops there), and R bends 60° before cracking. Write a model-to-measurement link for P.
Answer
The model predicts that a rigid pattern cannot slide, so the ceramic should break with very little bending. Ceramic P snapped at a bend of only 2°, compared with 60° for Metal R and at least 90° for Polymer Q.
This is consistent with the rigid-pattern model. The limit: one strip of each material was tested, so the result cannot show that all ceramics behave the same way.
What to study next
Continue with evaluating a fictional product claim, where you test a claim against measurements. The graph evidence and fair-comparison lab provides more data to link to models.
To practise writing the link with a teacher, see online one-to-one Science tuition.