Critical thinking skills
The skills of critical thinking are 6 specific, teachable operations, not a general quality of mind: mapping an argument, testing an inference, judging a source, reading a number, naming a failure, and auditing yourself. Each one has an object it works on, a question it asks, and an instrument that does the work. That is what makes them skills rather than a disposition, and it is why they can be practiced separately and assessed separately.
Defining critical thinking as "thinking about your thinking" is true and useless, because it gives nobody anything to do on Monday. The definition worth having is operational: critical thinking is the set of checks you run on a claim before you rely on it, and the 6 below are those checks.
The six skills and what each one does
| Skill | The question it asks | The instrument |
|---|---|---|
| Mapping an argument | What is the conclusion, and what is holding it up? | Premise and conclusion analysis |
| Testing an inference | Does the conclusion follow, and how strongly? | Validity for deduction, strength for induction |
| Judging a source | Who observed this, and what am I actually reading? | Provenance, primary against secondary |
| Reading a number | What is the denominator, and what is the baseline? | Base rates, percentages, axis and scale |
| Naming a failure | Which specific move went wrong here? | Fallacy vocabulary, used as a diagnosis |
| Auditing yourself | What did I want to be true before I started? | Disconfirmation search, bias checklists |
Mapping an argument
Mapping is the skill everything else waits on, because you cannot evaluate a claim you have not located. It means separating the conclusion from its support, marking each premise, and writing out the assumption nobody stated. Most disagreements that feel intractable turn out, once mapped, to be disagreements about a single unstated premise that neither side had said aloud. Logical reasoning covers the anatomy in full.
Testing an inference
Testing an inference means asking whether the premises reach the conclusion, and by what kind of route. A deductive route promises the conclusion cannot fail if the premises hold, so it is checked by looking for a case where the premises are true and the conclusion false. An inductive route promises only probability, so it is checked by counting cases and asking how they were chosen. Applying the deductive test to an inductive argument is the commonest technical error in amateur criticism, and Deductive vs inductive reasoning is the page that keeps the two apart.
Judging a source
Judging a source means establishing what you are holding: an observation, a report of an observation, or a summary of a report. Each step away from the original loses detail and adds interpretation, and the losses compound quietly. The concrete drill is to trace a claim back one link at a time until you reach the person or instrument that recorded it, and to notice how often the trail ends at a summary with nothing behind it.
Reading a number
Reading a number means finding what the number is a fraction of. A percentage without a denominator is not information, a change without a baseline is not a trend, and a chart with a truncated axis is a picture making a claim the figures do not support. This is the most learnable of the six, because it reduces to three habits: convert percentages to counts, ask how many were in each group, and read the axis labels before looking at the bars.
Naming a failure
Naming a failure means diagnosing the specific move that went wrong, and this skill comes with a warning attached. A fallacy label is not a rebuttal. Saying that an argument commits a named fallacy is the beginning of a criticism, not the end of one, and it obliges you to say what the argument needed and did not have. Labels used as verdicts become a way of dismissing arguments without reading them, which is a failure of critical thinking wearing its own vocabulary.
Auditing yourself
Auditing yourself means asking what you wanted the answer to be before you looked, then deliberately searching for the evidence that would disappoint you. It is the hardest of the six because it has no external object: nobody hands you a passage to work on, and the failure feels exactly like careful thought from inside. Confirmation bias is the page on the mechanism, and the practical countermeasure is procedural rather than attitudinal, which is to write down in advance what result would change your mind and then go looking for it.
Why these skills are important
These skills are important because each one blocks a specific, common failure, and the failures are expensive in ordinary work rather than only in philosophy seminars. Mapping prevents you from arguing with a position nobody holds. Inference testing prevents confident conclusions from unexamined general rules. Source judgment prevents a summary of a summary from being treated as a finding. Number reading prevents the small difference dressed as a large one. Failure naming turns a vague unease into a stated objection someone can answer. Self-auditing prevents the whole apparatus from being pointed only at conclusions you dislike, which is the way critical thinking most often goes wrong in practice.
Professional training builds the same six under local names. Nursing programs teach them as clinical judgment, engineering as hazard analysis and root cause investigation, auditing as professional skepticism. The vocabulary differs and the operations are identical, which is why a person who has learned them in one setting can usually transfer them to another faster than they can learn the new jargon.
How to build and enhance these skills
To build these skills, practice them one at a time against real material, and score yourself against a worked answer rather than against your impression. 5 steps that work:
- Pick one skill per week and run only that check, so the practice is deliberate rather than diffuse.
- Use material you did not choose, such as a report you were sent, since self-selected examples are the ones you already find easy.
- Write the answer out in full sentences, because a check performed silently cannot be reviewed.
- Compare against a worked solution, which is what Critical thinking exercises supplies for each of the six.
- Repeat the items you missed after a week, since recognition decays much faster than the sense of having understood.
Downloadable summaries and introductory guides written for beginners are a reasonable starting point for the vocabulary, and they cannot supply the skill, because none of the six is knowledge. Each is a procedure, and procedures are acquired by running them badly and being corrected. A reader who can define all six and has never worked an item has learned the names.
The check that tells you which skill you skipped
After you have reached a view on any contested claim, run this one:
Which of the 6 checks did I actually perform on this, and which did I skip because the conclusion already suited me?
The honest answer is usually that the checks were run thoroughly on the claim you disliked and lightly on the claim you found congenial. Noticing that asymmetry, on a specific claim, on a specific day, is worth more than another list of skills.