Bias and Limitations in Sources: Study Guide (English Language 6.3)

Why a writer needs to judge sources

Any argument that uses outside material takes on its strengths and weaknesses. Topic 6.2 showed that every source has a position and a perspective. This topic asks the next question: how far can a source be trusted to prove what you want it to prove? The course framework puts the writer’s responsibility this way: when incorporating evidence or sources into an argument, the strongest arguments recognise and acknowledge the biases and limitations of the material and account for those limitations in their reasoning.

This does not mean refusing to use imperfect sources. It means using them honestly: saying what they can show and what they cannot.

Bias

Bias is a leaning that distorts how a source presents its subject. It is not the same as having a view. A writer who argues for a policy has a position. A writer who ignores all evidence against it, uses loaded words and presents only favourable data shows bias.

The framework offers a useful measure: the degree to which a source does or does not consider other positions reflects the degree to which that source is biased. In other words, ask how fairly the source deals with what other people think.

Signs of low bias	Signs of high bias
States opposing views fairly and answers them	Ignores or caricatures opposing views
Uses measured language	Uses loaded or emotive language
Presents evidence that complicates its own claim	Presents only favourable evidence
Says how and by whom its data were gathered	Hides the method or the source of the data
Admits uncertainty (“suggests”, “may”)	Claims certainty beyond the evidence
No obvious stake in the outcome, or discloses it	Has an interest in the outcome and does not say

Vested interest matters. A company that sells filters has a reason to say tap water is poor; a water company has a reason to say it is fine. Neither is automatically lying, but each deserves a careful reading.

Bias also depends on purpose. The Declaration of Independence (1776) lists grievances against the King and says: “To prove this, let Facts be submitted to a candid world.” The list is deliberately one-sided because the document is an indictment, and it does not pretend to weigh the King’s side. A reader should judge a one-sided source in the light of what it is for, and should not treat it as a neutral report.

Limitations

A limitation is a feature of the evidence that restricts what it can prove. Even an unbiased source has limits. Common ones:

Limitation	Question to ask	Example
Sample size	How many people or cases?	“9 in 10” of 50 people
Sample selection	Who was asked, and how were they chosen?	Visitors to a company’s own stand
Scope	What does it cover and what not?	One region, one season
Date	How recent is it?	A report from before a major change
Method	How was it measured?	Self-reported taste, not blind tests
Type of evidence	Anecdote, expert opinion, data?	One person’s story as proof of a general trend
Correlation	Does it show cause or just association?	Two things that rose together

A limitation does not make evidence worthless; it tells you how far it goes. A small survey can suggest a trend without proving it. An anecdote can illustrate a point but cannot prove that the point is common.

Acknowledging and accounting for limitations

Recognising a bias or limitation is the first step. The second is to account for it in your reasoning. You can do this in three ways:

1. Qualify the claim. Say “suggests”, “may indicate” or “in this case” where the evidence supports no more.
2. Say what the limitation is. “Although the survey was small…” or “Because the brochure is produced by a company that sells filters…”.
3. Strengthen with other evidence. Pair a limited source with another that covers its gap, and say that you are doing so.

Compare:

Weak: A survey shows that 9 in 10 people prefer filtered water, so filters are better than tap water.

Strong: A filter company reports that 45 of 50 visitors to its own stand preferred the filtered water, but a company’s own visitors are a self-selected and interested group; the claim is better tested by the consumer group’s blind taste tests, where the difference was small.

The strong version names the source’s interest, describes the limitation of the sample and brings in independent evidence.

Worked example 1: evaluating and using a source

Source G (brochure from a water-filter company, fictional): “In a survey of 50 visitors to our stand at the town fair, 9 in 10 said our filtered water tasted better than tap water. Tap water is full of impurities that only our filter removes. Isn’t it time your family drank clean water?”

Task. Evaluate Source G and write a sentence that uses it responsibly in an argument that households do not need filters.

Step 1: Identify the position and perspective. Position: households should buy filters. Perspective: a company that sells them.

Step 2: Look for bias.

• Interest: the company profits from sales.
• Loaded language: “full of impurities”, “clean water” (implying that tap water is not clean).
• Rhetorical question: “Isn’t it time your family…?” appeals to fear for one’s family.
• Other positions: none considered.

Step 3: Look for limitations.

• Sample: 50 people.
• Selection: visitors to the company’s own stand, who are likely to be interested in filters.
• Method: a taste preference, not a safety test.
• Claim versus evidence: “full of impurities” is not supported by the survey, which concerned taste.

Step 4: Decide the use. The source is useful as evidence of what the company claims, not as proof about water quality.

Step 5: Draft a responsible sentence.

A filter company advertises that 45 of 50 visitors to its stand preferred its water, but the survey was small, drawn from people already visiting the company’s stand, and measured taste only, so it says little about whether tap water is unsafe.

Step 6: Check. The sentence names the source, states the claim accurately (45 of 50 is 9 in 10), identifies the interest and limitations and states what the evidence cannot show.

Worked example 2: comparing sources by degree of bias

Two more fictional sources on the same issue.

Source H (report by an independent consumer group, fictional): “We tested tap water samples from 120 homes in the region and compared them with three brands of home filter. All samples met the national safety standard. Filters improved taste scores in blind tests by a small margin, but the difference was not significant for most testers. The tests were carried out over two weeks in spring.”

Source I (letter from the regional water company, fictional): “Our water is tested daily and meets all safety standards. Complaints about taste are rare. We do not recommend that households buy filters for safety reasons.”

Step 1: Apply the measure of other positions. Source H reports the filters’ small benefit alongside its finding that tap water met standards, so it considers the case for filters fairly. Source I does not mention that some customers might prefer filtered water and gives no data beyond assertion.

Step 2: Consider interests. Source H is independent of both sellers and the water company. Source I comes from the water company, which benefits if households trust its water.

Step 3: Consider limitations. Source H: 120 homes in one region over two weeks in spring; the finding may not hold in other seasons or places. Source I: no figures, so the claim “complaints are rare” cannot be checked.

Step 4: Judge the degree of bias. Source H shows the least bias of the three: it presents evidence that cuts against a simple claim and gives its method. Source I shows more, with an interest and no evidence. Source G shows the most.

Step 5: Write the analysis sentence. Source H is the most reliable because it reports findings that complicate any simple conclusion and states its method, whereas Source I rests on an assertion from an interested party and Source G on a small, self-selected survey from a company that sells the product.

Common misconceptions

• “Bias means a source is lying.” Bias distorts emphasis and omission. A biased source can still contain true statements.
• “If I use a biased source, my argument is weak.” Using it without acknowledgment is weak; using it with acknowledgment can be strong.
• “Large numbers mean reliable evidence.” A large sample can still be biased if it is not representative.
• “A neutral-sounding source is unbiased.” Tone is not enough; check interests, omissions and method.
• “Limitations mean I should drop the evidence.” Qualify it, and pair it with other evidence.
• “A source’s bias is a reason to ignore its claims.” Assess each claim on its evidence.

Where this leads

Once you can judge the reliability of material, you may need to change your own claim. Topic 6.4 looks at revising a thesis for new evidence. To review how to combine sources, see Topic 6.1. Check your learning with the practice questions, the revision notes and the checklist.