United States
Hong KongEvaluating an argument: relevancy, sufficiency, validity
Skill 2 of Competency 3 reads "Evaluate specific claims in text based on relevancy, sufficiency, and validity of reasoning". Three separate tests. Relevancy asks whether the evidence bears on the claim, sufficiency whether there is enough, validity whether the conclusion follows.
The rule hands you the marking scheme. Three tests, named in the skill statement, and an argument can pass two and still fail.
The three tests in order
| Test | The question | Typical failure |
|---|---|---|
| Relevancy | Does this evidence bear on this claim? | A statistic about a different population or period |
| Sufficiency | Is there enough of it? | One anecdote supporting a general claim |
| Validity | Does the conclusion follow? | Correlation presented as cause |
Run them in that order, because relevancy is fastest to check and disposes of the most options. If the evidence is about something else, nothing downstream matters.
The reasoning failures that actually appear
- Correlation treated as causation. Two things moved together, so one caused the other.
- Sample too small or unrepresentative. One district, one year, one class.
- The comparison group is missing. Scores rose, but rose compared with what?
- Circular support. The evidence restates the claim in different words.
The missing comparison group is the one to watch for in education passages, because school data almost always needs a baseline and passages almost never state one.
A passage argues a summer program improved outcomes: Students who attended the summer program scored higher in the fall than those who did not. Which most weakens the argument?
- The program cost more per student than the district expected.
- Attendance at the program was voluntary, and students who signed up had higher scores the previous spring.
- A neighboring district runs a similar program.
- Some students who attended missed several sessions.
Strengthen and weaken items
The same three tests answer both directions. A weakener attacks the link between the evidence and the claim; a strengthener closes a gap in it. Neither one is usually a new fact about the topic, and options that add topic detail without touching the reasoning are the standard wrong answers.
Ask one question of every option: does this change how well the evidence supports the claim? If the answer is no, it is a distractor however interesting it is.
If Reading is the subtest you failed
Be realistic about what a retake changes. Reading has the highest published cut of the three multiple-choice subtests, at most 70% correct for a scaled 200, which is about 21 of 30. It also has the smallest gap between first attempt and best attempt: 78% and 86% in 2025, eight points, against fifteen for English Language Skills.
That combination says the same thing twice. Reading is not a memorization failure and sitting it again with the same habits is unlikely to change the outcome. Argument evaluation is one of the few parts of this subtest that responds directly to a method rather than to years of reading, which is why we would start here.
A retake is a fresh single-subtest registration at USD 32.50, and Rule 6A-4.0021(3)(d) requires at least thirty days since your previous administration of that subtest.
What we would practice
Take any opinion column. Write the claim in one sentence, list the evidence in bullets, then decide which of the three tests it fails. Most columns fail sufficiency. Doing this ten times is worth more than reading fifty practice questions, because it forces you to produce the judgment instead of recognizing it in someone else's.
Common questions
What are the three tests for a claim on the FTCE?
Relevancy, sufficiency and validity of reasoning, named in skill 2 of Competency 3. Relevancy asks whether the evidence bears on the claim, sufficiency whether there is enough of it, and validity whether the conclusion actually follows.
How do I answer a weaken-the-argument question?
Attack the link between the evidence and the claim rather than the topic. An option that raises cost, mentions a similar program elsewhere, or adds detail without touching the reasoning is a distractor, however plausible it sounds.
What is the most common reasoning error in these passages?
Correlation presented as causation, usually with no comparison group. Education passages report that something rose after a change was made, and the missing question is what it rose against. Watch for a baseline the passage never supplies.
Does retaking the Reading subtest usually work?
Less reliably than the others. The published 2025 figures are 78% at first attempt and 86% at best attempt, an eight-point gain against fifteen for English Language Skills and the essay. A retake without a change of method is a poor bet here.