A model prediction says what should happen if the assumptions hold. A documented outcome says what was recorded once real events, with real changes, took place. Keep them in separate sentences and never use one as proof of the other.
This lesson closes the main sequence in government and the economy. It builds on tracing a policy mechanism using a stated model, which produces the prediction you test here.
Why are prediction and outcome different things?
A prediction holds everything else constant by assumption. An outcome has no such protection, because other things change in the real world. A match is therefore encouraging evidence, not proof.
Worked example: a fictional tax on soft drinks
Zenia, an invented country, puts a tax of RM2 on each bottle of a soft drink. Before the tax, the price is RM5 and sales are 10 000 bottles a week. After the tax, the price is RM6.40 and sales are 8 500 bottles.
Prediction. The tax shifts supply left, so the model expects a higher price and a lower quantity.
Outcome. The price rose by RM1.40, which is 28% of RM5. Sales fell by 1 500 bottles, which is 15% of 10 000.
Tax share. Buyers paid RM1.40 of the RM2 tax, which is 70%. Sellers absorbed RM0.60.
The direction of both changes matches the prediction. Yet one more fact could change the reading: the tax started in the school holidays, when sales of the drink could have been different anyway.
What does the match show?
It shows that the data is consistent with the model. It does not show that the tax alone caused the fall, because the holiday is a second possible cause.
| Sentence | Claim | Verdict |
|---|---|---|
| “The data proves the model is correct” | The match settles the matter | Overclaims |
| “The data is consistent with the model” | The match supports the prediction | Careful |
| “The tax caused the whole 15% fall” | One cause only | Not shown |
| “Part of the 15% fall may come from the tax” | One possible cause | Careful |
The mistake to avoid
The common slip is treating one matching data point as proof. Compare the endings of two answers.
Wrong: “Sales fell after the tax, so the model is proved and the tax works.”
Right: “Sales fell 15% after the tax, which is consistent with the model’s prediction. Other changes, such as school holidays, may also have affected sales, so the data does not isolate the tax.”
The right answer also avoids “works”, a judgement word. It describes what was recorded and what it supports.
A three-sentence pattern
- Prediction: “The model predicts… if…”
- Outcome: “The data records…”
- Reading: “This is consistent with the prediction, but other factors such as… may also explain it.”
When the data disagrees, change step 3: “This differs from the prediction, which may mean that the assumption that… did not hold.” For more on qualifying claims with the data in front of you, see evidence-based economic answers.
Check yourself
After a subsidy on rice, the price fell from RM4.00 to RM3.60 per kilogram, and weather that year was unusually good for harvests. Write the reading sentence.
Answer
The fall is RM0.40, which is 10% of RM4.00.
“The 10% fall in price is consistent with the model’s prediction that a subsidy shifts supply right. However, good weather could also have raised supply, so the data does not show that the subsidy alone caused the fall.”
Check: the sentence says “consistent with”, names a second factor, and avoids “proves”.
What to study next
Practise the skill in the government and the economy practice set. Record any overclaiming in the mistake log so you can spot the pattern.
If you want a teacher to question each of your conclusions, see online one-to-one Economics tuition.