Lesson 19: Writing with Evaluation

Evaluation is what lifts a TOK essay from competent to compelling. It means making reasoned judgements about the strength, scope, limits and implications of claims, rather than only describing or listing them. Examiners look for evidence that you can weigh arguments, recognise assumptions, and draw justified conclusions. This lesson explains what evaluation looks like in writing, with sentence-level techniques and model paragraphs.

What counts as evaluation

Description: "Peer review is a process in which experts check research before publication." Analysis: "Peer review increases reliability because independent experts can detect errors and unsupported claims." Evaluation: "Peer review increases reliability, but its protection is partial: reviewers rarely repeat experiments, and the system can reward conformity, as the slow acceptance of Barry Marshall and Robin Warren's discovery that bacteria cause stomach ulcers suggests. Peer review is therefore best seen as a filter that reduces some kinds of error, not as a guarantee of truth." Notice how the evaluative version states a judgement, supports it with an example, identifies limits, and refines the claim.

A paragraph structure for evaluation

1. Claim: a clear topic sentence linked to the title. 2. Explain the knowledge idea: why does the claim hold? Which TOK concept or mechanism is at work? 3. Example: specific details that make the idea concrete. 4. Analyse the example: what does it show, exactly? 5. Counterclaim or limit: a respectful challenge to your own point. 6. Evaluate: weigh the two; say which is stronger in which conditions; refine the claim. 7. Link: show how the paragraph supports your thesis or the title.

Sentence stems for evaluation

• "This is persuasive because... but it depends on the assumption that..."
• "A limitation of this approach is... which suggests that..."
• "The strength of this evidence lies in... whereas its weakness is..."
• "This applies most clearly in contexts where..., but less so where..."
• "The implication is that..., especially for those who..."
• "While X supports the claim, Y complicates it, because..."
• "On balance, ... since..."

Avoid empty hedges ("it could be said") without reasons. Evaluation requires *because*.

A model paragraph (about 280 words, illustrative)

*Title idea: "To what extent can we trust experts?"* "Trust in expert knowledge is strongest when claims are exposed to systematic checking. In the natural sciences, the detection of gravitational waves announced in 2016 was based on signals recorded by two detectors thousands of kilometres apart, analysed by an international collaboration of over a thousand scientists, and later confirmed by further detections. The independence of the instruments and the number of analysts make it difficult for a single bias or error to produce the result, which gives non-specialists good reason to accept the claim even though they cannot verify the mathematics.

This kind of trust, however, is a trust in a *system* rather than in individual authority. The system has limits. Funding pressure can influence which questions are pursued, and the reward for novel findings has contributed to reproducibility problems in some fields. Moreover, the confidence that attaches to a well-tested physical measurement cannot simply be transferred to forecasts about complex social systems, where the reflexivity of human behaviour and the difficulty of controlled experiments reduce predictive power.

Thus trust in experts is reasonable to the extent that the field has transparent methods, independent verification and a record of successful prediction. The rational response is therefore proportionate trust: high for the existence of gravitational waves, more cautious for long-range economic forecasts. This graded approach avoids both uncritical deference and blanket scepticism, and it suggests that the question for knowers is not 'Do I trust experts?' but 'What features of this expertise justify trust, and how can I check them?'"

Techniques that deepen evaluation

• Compare areas of knowledge to show how methods affect reliability.
• Examine assumptions hidden in claims or definitions.
• Consider scope: under what conditions does the claim hold?
• Explore implications: what follows for individuals, communities or policy?
• Use counterexamples that genuinely test your claim.
• Distinguish degree from kind: it is rarely "all or nothing".
• Acknowledge disagreement within a field and ask what would count as evidence for each position.

Common weaknesses

• List-like paragraphs: several examples with one sentence each and no weighing.
• Assertion without reasons: "Clearly, science is more reliable than art."
• Strawman counterclaims that are too weak to challenge anything.
• Conclusions that merely repeat. A conclusion should advance by explaining what the arguments, taken together, show.
• Overuse of absolute terms (always, never, impossible) without justification.

Editing for evaluation

Read each paragraph and highlight the evaluative sentences. If a paragraph has none, revise. Ask "So what?" after each key sentence. Replace general nouns with specific ones. Cut descriptive detail that is not used in the analysis. Check transitions: does each paragraph lead to the next?

Case file: improving one paragraph in three stages

Stage 1 (descriptive): "In history, historians use sources. Primary sources are from the time. They can be letters or diaries. Historians also use secondary sources. These are written later." Stage 2 (analytical): "Historians rely on primary sources, such as letters and diaries, because they were produced close to events. However, a letter writer sees only part of the situation, so historians compare several sources." Stage 3 (evaluative): "Historians prefer primary sources because they were produced close to events, but proximity does not guarantee accuracy. The diaries of an officer in the 1857 uprising, for instance, record what he saw and feared, and omit the experience of local people whose voices appear mostly in oral traditions and later petitions. Historians therefore weigh the diaries against these other sources, using their differences to identify the limits of each. The reliability of historical knowledge thus emerges from the comparison of imperfect sources, not from any single document." The movement from stage 1 to stage 3 is the movement from list to explanation to judgement. A useful test: could a reader tell what you *think* from your paragraph?

Two sides of a question

"Does using sources from several perspectives make an argument weaker or stronger?"

• Weaker: Too many perspectives can scatter the argument and leave no clear position.
• Stronger: Considering other perspectives tests your claim, and a conclusion that survives challenge is better supported.

The remedy for scattering is to use perspectives to *serve the thesis*: show how they refine or limit it.

Evaluative moves to practise

• Weigh: "While A is supported by x, B is supported by y; on balance A is more persuasive here because z."
• Qualify: "This holds when the sample is large, but not for small groups."
• Extend: "If this is true, it implies that..."
• Challenge assumptions: "The argument assumes that..., which may not hold in..."
• Contrast: "In the natural sciences..., whereas in the arts..."

Glossary for this lesson

• Evaluation: a reasoned judgement about the strength or limits of a claim.
• Qualification: a condition that limits how far a claim applies.
• Implication: a consequence that follows from a claim.
• Warrant: the reasoning that links evidence to a claim.
• Nuance: attention to degrees and conditions rather than all-or-nothing claims.

Discussion questions with guidance

• *How long should an evaluative paragraph be?* Usually 200–350 words, enough to introduce an idea, support it with an example, and weigh it against a counterclaim.
• *Can a conclusion contain evaluation?* Yes, it should: it weighs the arguments to answer the title, and explains implications.
• *What if the evidence is genuinely balanced?* Say so, and explain what additional evidence or condition would tip the balance. This is itself a form of evaluation.
• *Is it better to use formal language?* Prefer clear, precise language over elaborate vocabulary. Simple sentences that express complex ideas are valued.

Practice: from description to evaluation

Take any paragraph you have written that is mostly description. Underline every sentence that begins with "This shows", "This is because" or "However". If you cannot find them, add them. Then check that each "However" introduces a real limitation and is followed by a sentence that weighs it. Revising only for these moves often changes the quality of an essay significantly.

Self-test

Write two sentences about a source you use in another subject: one descriptive, one evaluative. Explain what makes the second evaluative.

Key points to remember

• Evaluation is weighing: it judges strength, scope and limits with reasons.
• Use a repeatable paragraph structure: claim, idea, example, analysis, counterclaim, evaluation, link.
• Replace absolutes with qualified, justified claims.
• Check each paragraph for at least two evaluative sentences, such as "This is persuasive because..., but it depends on...".
• Read your essay aloud to hear weak transitions and unsupported assertions.
• Ask "So what?" after every key sentence, and add the implication if it is missing.

Quick reflection

Count how many sentences in your latest paragraph describe, how many explain, and how many weigh. Aim to increase the proportion that weigh. Then reread the paragraph aloud and mark any sentence where you would ask, "How do you know that?" and answer it with evidence or a qualification.