Given an argument, how do we judge it fairly and systematically, and how do we assess the sources its premises rest on?
Apply a systematic method for evaluating an argument and assessing the reliability, relevance and bias of the sources its premises depend on
A focused answer on the systematic evaluation of arguments and sources. The two-question method (does it follow, are the premises true), assessing source reliability and bias, weighing counter-considerations, and reaching a justified verdict.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
Jump to a section
What this dot point is asking
SEAB wants you to bring together everything in the reasoning area into a single, repeatable procedure for evaluating an argument, and to add the skill of judging the sources its premises rest on. This is the core competence of the critical thinking paper. The output is always a justified verdict: not merely whether you agree, but whether the argument succeeds, with reasons that anyone could check.
The answer
The two-question core
Every argument evaluation reduces to two independent questions. First, does the conclusion follow from the premises? For a deductive argument this is the question of validity; for an inductive one it is the question of strength. Second, are the premises actually true? An argument succeeds only if both questions get a yes: the reasoning is good and the premises are true. Keeping the two questions separate is the discipline that prevents most evaluation errors.
A systematic method
A reliable procedure runs as follows. Reconstruct the argument in standard form, surfacing unstated assumptions. Classify it as deductive or inductive. Assess the inference (validity or strength), looking for formal and informal fallacies. Assess the truth of the premises, including the reliability of their sources. Weigh counter-considerations and the strongest objections. Then reach a justified verdict that follows from the assessment. Working in this order stops you from rejecting a conclusion you dislike on the wrong grounds.
Assessing source reliability
Premises often rest on sources, so evaluating them means assessing those sources. The main criteria are: expertise (is the source qualified in the relevant field?); track record (has it been accurate before?); independence (does it have an interest in the conclusion?); primary versus secondary status (first-hand evidence or a report of a report?); transparency of method (can the claim be checked?); and corroboration (do independent sources agree?). Reliability is a matter of degree assessed across these criteria, not a yes or no.
Bias is a reason to scrutinise, not to dismiss
A biased source, one with an interest in the conclusion, warrants extra scrutiny, but bias does not make a claim false. To reject a claim solely because of its source is the genetic fallacy, and to reject an argument solely by attacking the arguer is ad hominem. The correct response to bias is to look harder for independent corroboration and to check the method, not to dismiss the content outright.
Weighing counter-considerations and reaching a verdict
A fair evaluation states the strongest objection to the argument and the strongest reply, then judges where the balance lies. The verdict should be proportionate: an argument can be strong but not decisive, or fail on one premise while succeeding elsewhere. The mark of good evaluation is calibration, matching the confidence of the verdict to the strength of the case, rather than treating every argument as wholly right or wholly wrong.
Examples in context
Example 1. Two studies, two funders. Two studies reach opposite conclusions about a food additive; one is funded by the manufacturer, the other by an independent agency. A careful evaluator does not simply believe the independent one and dismiss the other, but scrutinises both methods, checks sample sizes and looks for further independent replication. The case shows that source assessment sharpens, rather than replaces, the evaluation of the evidence itself.
Example 2. A viral statistic. A striking figure spreads online with no original source cited. Applying the method, the figure is a premise resting on an untraceable source, so it cannot be assessed for expertise, method or independence. The correct response is to suspend judgement and seek the primary source, illustrating how the absence of a checkable source is itself a reason to withhold trust.
Try this
Q1. State the two independent questions at the core of argument evaluation. [6 marks]
- Cue. Does the conclusion follow from the premises (validity or strength)? And are the premises actually true (including the reliability of their sources)? Both must be answered yes for the argument to succeed.
Q2. List four criteria for assessing the reliability of a source. [6 marks]
- Cue. Any four of: relevant expertise, track record, independence (freedom from interest in the conclusion), primary versus secondary status, transparency of method, and corroboration by independent sources.
Q3. Explain why discovering that a source is biased does not by itself refute its claim. [8 marks]
- Cue. Bias is an interest in the conclusion, not evidence that the claim is false; rejecting a claim solely because of its source is the genetic fallacy. The right response is extra scrutiny and a search for independent corroboration.
Exam-style practice questions
Practice questions written in the style of SEAB exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
Original12 marksSet out a systematic method for evaluating an argument, and apply it to: 'A leading think tank reports that cutting corporate tax raised investment in three countries. So Singapore should cut corporate tax to raise investment.'Show worked answer →
A strong answer states a method: (1) reconstruct the argument in standard form; (2) classify it as deductive or inductive; (3) assess whether the conclusion follows (validity or strength); (4) assess whether the premises are true, including the reliability of their sources; (5) weigh counter-considerations; (6) reach a justified verdict.
Apply it. Reconstruct: Premise, a think tank reports that tax cuts raised investment in three countries. Conclusion, Singapore should cut corporate tax to raise investment. This is an inductive generalisation plus a policy recommendation.
Assess the inference: a three-country sample is small, the countries may differ from Singapore in relevant ways, and rising investment may have other causes (correlation not causation). So the inductive support is weak.
Assess the premises and source: a think tank may have a policy agenda, so check for funding bias and whether the study is peer-reviewed or cherry-picked. Reliability is uncertain without more detail.
Weigh counter-considerations: even if investment rose, the policy has costs (lost revenue) the argument ignores, so the move from a causal claim to "should" needs an unstated evaluative premise.
Verdict: the argument is weak as it stands. Markers reward an explicit method, correct application, the small-sample and source-bias points, and a decided conclusion.
Original12 marksHow should we assess the reliability of a source when evaluating an argument? Discuss with examples.Show worked answer →
The expected answer offers criteria for source assessment: expertise (is the source qualified in the relevant field?), track record (has it been accurate before?), independence and bias (does it have an interest in the conclusion?), primary versus secondary status (is it first-hand evidence or a report of a report?), corroboration (do independent sources agree?), and transparency of method (can the claim be checked?).
Apply with examples: a peer-reviewed study by qualified researchers with declared funding scores higher on expertise and transparency than an anonymous social media post; a body funded by an industry reporting a result that benefits that industry warrants extra scrutiny for bias, though bias does not automatically make a claim false.
Make two careful points. First, bias is a reason to scrutinise, not to dismiss; a biased source can still be right, so attacking only the source risks the ad hominem and genetic fallacies. Second, corroboration by independent sources is one of the strongest reliability signals.
Judgement: reliability is a matter of degree assessed across several criteria, and the right output is calibrated trust, not blanket acceptance or rejection. Markers reward a set of criteria, examples, the bias-is-not-refutation point, and the emphasis on corroboration and method.
Related dot points
- Explain deductive validity and soundness, distinguish them from the truth of the premises, and apply the concepts to assess given arguments
A focused answer on deductive arguments. What validity is (truth-preserving form), why it differs from the truth of the premises, what soundness adds, and how to test a deductive argument for both.
- Distinguish inductive from deductive reasoning and assess inductive strength across generalisation, analogy and inference to the best explanation
A focused answer on inductive reasoning. How induction differs from deduction, what makes an inductive argument strong or weak, and the main forms: enumerative generalisation, analogy, and inference to the best explanation.
- Identify the conclusion, premises and unstated assumptions of an argument and represent its structure, distinguishing argument from non-argument
A focused answer on argument reconstruction. Finding the conclusion and premises, spotting indicator words, surfacing unstated assumptions, distinguishing arguments from explanations and assertions, and mapping argument structure.
- Identify and explain common formal and informal fallacies and diagnose them in given arguments without committing the fallacy-fallacy
A focused answer on fallacies. The difference between formal and informal fallacies, a working catalogue (affirming the consequent, ad hominem, straw man, false dichotomy, slippery slope, equivocation, appeal to authority and others), and how to diagnose them fairly.