Separating Evidence from Values in a Policy Debate

By Better Wise

Where This Lesson Fits

Module 2 trained you to ask who made a claim, what the source was, and whether the numbers survived a baseline check. That skill catches bad information. It does not, on its own, let you evaluate a good argument that you disagree with. This lesson adds the missing move: splitting a policy argument into the part evidence can test and the part it cannot. When you finish, you should be able to:

• Separate any policy sentence into its factual claims and its value judgments.
• Sort claims into three buckets: evidence settles it, evidence informs it, evidence cannot settle it.
• Spot value judgments wearing the costume of facts ("too expensive", "common sense", "obviously").
• Explain how two people can accept identical evidence and still disagree honestly.

Every Policy Sentence Carries Two Payloads

A factual claim is a statement about the world that evidence can check: what happened, how much something costs, who is affected, what a policy did in another city.

A value judgment is a statement about which priority should win when good things conflict: fairness, safety, cost, freedom, convenience, the environment.

Neither type is inferior. A debate with only facts has no direction, and a debate with only values has no contact with reality. The trouble starts when one is dressed as the other.

Worked example 1: one headline, two payloads

"City bus lanes cut car traffic 12% but cost $4.2 million a year."

• "Traffic fell 12%" is factual. Ask: measured how, over what period, compared to what baseline?
• "Cost $4.2 million a year" is factual. Ask: gross or net, and what is included?
• The word "but" is doing value work. It ranks the cost above the traffic drop without saying so. Swap it for "and" and the same facts read differently.

Notice that nobody lied. The framing carried a value that never got stated out loud. Your job is to hear it.

Worked example 2: a full argument, dissected

"The city should charge 10 cents for single-use bags. Plastic bags take centuries to break down and clog drains, which contributed to flooding last spring. In neighbouring towns, a similar charge cut bag use by 70%. Some shops will lose customers, but the environment matters more than convenience."

The strongest version of this argument keeps the factual rows checkable and says the last row out loud. That last sentence is not a weakness. It is the part a reader can actually respond to.

Worked example 3: same evidence, different values

Two residents read the same report. Both accept the 70% drop and the $4.2 million price. They still disagree, and neither is being irrational.

• One ranks flood risk for households in low-lying streets above the cost to shops.
• The other ranks the survival of small retailers above a modest reduction in drain blockages.

Here is the useful part: evidence can inform a tradeoff, but it cannot pick the winner. Evidence can still narrow the fight enormously. It can tell you which streets flood, how often, how many shops are at risk, and who actually pays the charge. After that, the remaining disagreement is smaller and honest.

The three-bucket sort

flowchart TD
    A["🗣️ A claim in a debate"] --> B{"Can it be measured or checked?"}
    B -->|Yes| C["📊 Factual claim: go get the evidence"]
    B -->|No| D{"Is it about which priority should win?"}
    D -->|Yes| E["⚖️ Value judgment: state it openly"]
    D -->|No| F["🔍 Definition problem: pin down the words first"]
    C --> G["Does the policy actually deliver the value it was sold on?"]
    E --> G

The third bucket matters more than beginners expect. Many "factual" claims are really definition problems. "Is this expensive?" needs a comparison group before it is even a measurable question.

Disguised values

Watch for these, because they pass as neutral:

• "Too expensive" hides a value. "Too" compared to what, and worth less than what benefit?
• "Wasteful" assumes the spending has no value, rather than arguing it.
• "Common sense" asks you to accept an unstated assumption without testing it.
• "Unfair to low-income households" is genuinely mixed: a factual part (who pays, and what share of their budget) and a value part (what fairness requires).

Split the mixed claim and both halves become workable. Leave it whole and the argument goes nowhere.

Habits That Keep Facts and Values Apart

• Ask "what number, measured how, compared to what?" whenever a claim sounds factual. This is the same discipline from Module 2, aimed one step earlier.
• Ask "which priority wins?" whenever a claim sounds like an insult or a slogan.
• Do not use "that is just your values" as a put-down. Everyone has values, including you. The problem is a hidden value, not a value.
• Run the means-ends check: your value is the destination, your policy is the vehicle. If the vehicle does not reach the destination, change the vehicle. Do not abandon the destination to defend a bad policy.
• Write the two columns down before you argue. Ten seconds of sorting saves an hour of talking past each other.

What to Take Away

• Most policy sentences carry a factual claim and a value judgment at the same time.
• Sort claims into three buckets: evidence settles it, evidence informs it, evidence cannot settle it.
• Naming a value does not weaken your case. It shrinks the disagreement to the part that is actually contested.
• Check whether your preferred policy delivers the value you hold, and be willing to swap the policy.

Next, you will take the value you have now made explicit and stress-test it, by building the strongest version of the argument against it before you respond.

Front: Factual claim Back: A statement that can be checked against evidence: measurements, records, data.

Front: Value judgment Back: A statement about which priority should win, or what is fair, right, or acceptable.

Front: What are the three buckets for sorting a policy claim? Back: Evidence can settle it, evidence can inform it, evidence cannot settle it.

Front: What kind of claim is 'the plan is too expensive'? Back: A value judgment wrapped around a number, because 'too' depends on what the money is worth.

Front: What kind of claim is 'this charge is unfair to low-income households'? Back: A mixed claim: a factual part (who pays) plus a value part (what fairness requires).

Front: Two people accept the same data but still disagree. What is the usual reason? Back: They rank competing values differently, such as flood protection versus shop costs.

Front: What is the means-ends check? Back: Ask whether the policy actually delivers the value it was sold on. If not, change the policy.

Front: Why is 'common sense' a warning sign in a debate? Back: It presents an unstated value or assumption as if it needed no evidence.

Front: Best question to test a factual claim? Back: What number, measured how, and compared to what?

Front: Best question to test a value claim? Back: Which priority wins when two good things conflict?

Front: What can evidence do to a value disagreement? Back: It cannot prove a value, but it can show whether a policy serving that value works.

Front: Does naming your value weaken an argument? Back: No. It makes the disagreement honest and shrinks it to the part that is really contested.

Front: In the bus lane headline, what work was the word 'but' doing? Back: It quietly ranked the cost above the traffic drop without stating that ranking.

1. A statement that can be checked against data is a _____, while a statement about which priority should win is a _____.
2. The phrase "the plan is too expensive" hides a _____ inside a number, because "too" depends on what you think the benefit is worth.
3. Evidence can _____ a tradeoff, but it cannot _____ which side should win.
4. When a policy fails to deliver the value it was sold on, the honest response is to change the _____, not the value.