Table of Contents
- 1. Lesson Overview
- 2. Learning Outcomes
- 3. Before You Begin
- 4. Unit 10 Retrieval Practice
- 5. Opening Scenario
- 6. Core Content
- 6.1 Information vs. Evidence
- 6.2 Source vs. Citation
- 6.3 Primary, Secondary, Tertiary Sources
- 6.4 Authority and Expertise
- 6.5 Relevance, Scope, and Context
- 6.6 Currency and Version
- 6.7 Provenance and Transformations
- 7. CLAIM Walkthrough
- 8. Verification-Process Walkthrough
- 9. Worked Example
- 10. Guided Learning Activities
- 11. Exercises
- 12. Homework Problems
- 13. Mini-Project
- 14. Self-Check Quiz
- 15. Assignment Connections
- 16. Reflective Journal Prompts
- 17. Key Takeaways
- 18. Key Terms
- 19. Looking Ahead
- 20. Sources & Further Reading
1. Lesson Overview
This lesson establishes the foundational concepts you will use throughout Unit 11 to verify claims—especially claims generated by or assisted by AI. You will learn to distinguish information from evidence, a source from a citation, and to classify sources as primary, secondary, or tertiary depending on your verification question.
You will also develop a systematic approach to evaluating whether a source is authoritative, relevant, and sufficiently current for a particular claim. These evaluations are not checkboxes; they depend on the claim, the field, the context, and the consequences of being wrong. By the end of this lesson, you will be able to select the most appropriate source type for a claim and justify your choice using evidence-focused reasoning.
This lesson prepares you for Lesson 11.2 (claim-evidence alignment and corroboration), for the cumulative Unit 11 project, and for Assignment 5, where you will verify an entire AI-generated passage.
2. Learning Outcomes
By the end of this lesson, you should be able to:
- Define information, evidence, source, citation, authority, relevance, currency, provenance, primary source, secondary source, and tertiary source.
- Distinguish information from evidence and explain why information becomes evidence only in relation to a claim.
- Distinguish a source from a citation and explain why a citation does not prove the source exists, supports the claim, or is authoritative.
- Classify sources as primary, secondary, or tertiary in context, and explain why the same source may be primary for one question and secondary for another.
- Evaluate source authority relative to a specific claim, considering expertise, mandate, method, transparency, and conflicts of interest.
- Evaluate source relevance, scope, population, jurisdiction, period, method, and context.
- Evaluate source currency and version, and determine when an older source remains appropriate.
- Identify provenance and possible transformations of information as it moves between sources.
- Select the strongest source type for claims involving law, policy, science, history, software, statistics, and professional recommendations.
- Justify source selection and evaluation using the concepts of authority, relevance, currency, and provenance.
3. Before You Begin
- Assigned reading: Chapter 8 of the course textbook: Verification, Evidence, and Epistemic Responsibility (sections on evidence, sources, authority, and relevance).
- Prerequisite knowledge: You should be able to identify claims, decompose compound claims, distinguish hallucination from fabrication, and apply the TRACE framework (Unit 10).
- Library and search tools: You may use your institution’s library portal, public search engines (e.g., Google Scholar, Internet Archive), or other open-access resources. No paid database is required.
- Optional AI tool: You may use an approved AI system for discovery support (e.g., suggesting search terms), but you must independently verify every source. AI-generated answers are not final authority.
- Supplied-source alternative: Every activity in this lesson includes a fully supplied source set so that you are never disadvantaged by lack of access.
- Privacy and copyright: Do not upload confidential, proprietary, or third-party material to any unapproved system. Use only synthetic, public, or supplied information.
- Academic integrity: All work you submit must be your own. If you use AI for discovery, you must document that use and verify every claim independently.
- Estimated preparation time: 20–30 minutes to review Chapter 8 and the Unit 10 summary.
4. Unit 10 Retrieval Practice
The following questions briefly review TRACE and claim analysis from Unit 10. Answer each before revealing the feedback.
Question 1: Read this synthetic passage and identify two claims, one possible fabrication, one unsupported claim, and one uncertainty:
“A 2021 study by the Global Health Institute found that 78% of adults who consumed more than 400 mg of caffeine daily experienced improved memory recall. The lead author, Dr. Elena Rivas, stated that caffeine ‘fundamentally rewires the neural pathways responsible for long-term retention.’ However, a 2020 review suggested that the effect may be limited to certain age groups. Caffeine is the most widely used psychoactive substance in the world.”
Then prioritize one claim for verification, explaining why it is the highest priority.
Show retrieval practice feedback
Two claims:
- Claim A: “78% of adults who consumed >400 mg caffeine daily experienced improved memory recall” (from the 2021 study).
- Claim B: “Caffeine fundamentally rewires neural pathways responsible for long-term retention” (attributed to Dr. Rivas).
Possible fabrication: The phrase “fundamentally rewires the neural pathways” is strong and mechanistic; it may be a fabricated or exaggerated quotation. A neuroscience claim of that magnitude would require direct evidence from imaging or electrophysiology, which the passage does not describe.
Unsupported claim: “Caffeine is the most widely used psychoactive substance” may be true, but the passage provides no source for it.
Uncertainty: The 2020 review suggests the effect “may be limited to certain age groups” – this is a qualification that the original claim omits.
Priority: The highest-priority claim for verification is the 78% statistic. If this number is wrong, the entire conclusion about caffeine and memory is undermined. It is also a quantitative claim with a specific population and measure, making it verifiable against the original study.
Question 2: Using the TRACE framework, what type of error is most likely present in the claim “caffeine fundamentally rewires neural pathways”? Choose from: hallucination, fabrication, factual error, unsupported claim, omission, contextual mismatch, invalid inference, or false precision.
Show answer
Answer: The claim is likely an unsupported claim with potential fabrication (if the quotation is invented) or invalid inference (if the study only showed an association but the quotation states causation). The word “fundamentally rewires” goes far beyond what a single study could establish.
Question 3: Why is it insufficient to ask the AI system that generated a claim whether that claim is correct?
Show answer
Answer: An AI system does not have independent access to evidence; it generates responses based on patterns in its training data. Asking the same system to “check itself” does not introduce new information, and the system cannot evaluate source authority, relevance, or currency in an accountable way. Verification requires external evidence and human judgment.
5. Opening Scenario: A Polished AI Response
This scenario runs throughout the lesson. Read it carefully and return to it as you learn each new concept.
You are a policy analyst in a provincial ministry of education. You ask an AI assistant for a summary of research on the effectiveness of “blended learning” (combining online and in-person instruction) in secondary schools. The AI produces this response:
“A 2022 meta-analysis by the Centre for Educational Innovation found that blended learning improves student outcomes by an average of 18% compared to traditional instruction. According to Dr. Mark Harrison in a 2020 review, blended learning ‘is the most effective pedagogical approach of the last fifty years.’ The Ministry of Education in British Columbia implemented blended learning in all Grade 10 classrooms in 2021, reporting a 12% increase in graduation rates. For a practical overview, see the EdTech Today blog post ‘Blended Learning in Action’ (2023).”
The AI response cites:
- A 2022 meta-analysis by the Centre for Educational Innovation
- A 2020 review by Dr. Mark Harrison
- A 2021 British Columbia Ministry of Education policy change
- A 2023 blog post from EdTech Today
Your task: Before you verify anything, ask yourself:
- Which of these are primary, secondary, or tertiary sources?
- Which source is most appropriate for each claim?
- Does the authority of each source depend on the claim being made?
- What information is missing that you would need to evaluate these sources?
Show guided analysis of the scenario
1. Source types (provisional):
- Meta-analysis: A secondary source (it synthesises primary studies).
- Review by Dr. Harrison: Secondary source (interpretive, synthetic).
- BC Ministry policy: If the policy document itself is cited, it is a primary source (official government record).
- Blog post: Tertiary or secondary, depending on content; likely a summary of other sources.
2. Most appropriate source for each claim:
- The 18% improvement claim should be verified against the original meta-analysis (if it exists).
- The quotation should be traced to Dr. Harrison’s original review or transcript.
- The policy claim should be verified against the official BC Ministry document (not a news summary).
- The blog post is not authoritative for any of the substantive claims; it may be useful for orientation only.
3. Authority depends on the claim: The meta-analysis authors may be authoritative for statistical synthesis, but they are not authoritative for BC policy. The Ministry is authoritative for its own policy but not for pedagogical psychology. Dr. Harrison may be authoritative in education research, but the quotation is extreme and needs context.
4. Missing information:
- Full bibliographic details (authors, titles, journals, DOIs, URLs).
- Whether the meta-analysis was peer-reviewed.
- Whether Dr. Harrison’s review is peer-reviewed or opinion.
- Whether the BC policy change actually occurred and affected all Grade 10 classrooms.
- How the 18% and 12% figures are defined (what was measured, what populations, what time frames).
6. Core Content
6.1 Information vs. Evidence
Information consists of data, statements, observations, records, documents, representations, or communications. Evidence is information used to support, qualify, challenge, or evaluate a particular claim.
The same piece of information may be evidence for one claim and irrelevant for another. The word “78%” is just a number (information) until it is used to support a claim such as “most adults prefer online learning.” To become evidence, information must be connected to a claim in a way that is relevant, verifiable, and accountable.
Why this matters for AI verification: AI systems generate fluent responses filled with information. But information is not automatically evidence. You must ask: What claim is this information supposed to support? Does it actually do so?
Consider the statement “The Earth orbits the Sun.” Is this information or evidence? Explain.
Show feedback
Answer: It is information until it is used in relation to a claim. For the claim “Copernicus was correct,” it is evidence. For the claim “Jupiter is the largest planet,” it is irrelevant information. The classification depends on the claim.
6.2 Source vs. Citation
A source is the person, record, dataset, document, publication, institution, observation, or other origin from which information is obtained. A citation identifies or points to a source.
Citations are references, not proof. A citation does not prove that:
- the source exists;
- the metadata are correct;
- the source was read;
- the source supports the claim;
- the source is authoritative;
- permission exists;
- the evidence is current.
AI systems often generate citations that look real but are fabricated, misattributed, or irrelevant. You must verify the citation and the source.
An AI response includes “(Smith, 2019, p. 42).” List three things you need to verify beyond the existence of the citation.
Show feedback
Possible answers:
- Does Smith (2019) exist (is the reference real)?
- Does page 42 actually contain the claim?
- Does Smith (2019) support the claim as stated, or has the claim been misrepresented?
- Is Smith an authoritative source for this specific claim?
6.3 Primary, Secondary, and Tertiary Sources
Primary source: Provides direct, original, or first-hand evidence relevant to the field or question.
Secondary source: Interprets, analyses, critiques, compares, or synthesises primary material.
Tertiary source: Summarises or organises information drawn from primary and secondary sources.
| Source type | Description | Examples |
|---|---|---|
| Primary | Direct, original evidence | Original research article, statute, policy document, dataset, court decision, historical letter, interview transcript, technical standard |
| Secondary | Interpretation or synthesis of primary sources | Review article, scholarly analysis, legal commentary, systematic review, critical essay, meta-analysis |
| Tertiary | Summary or organisation of primary and secondary sources | Encyclopedia, introductory textbook, general reference page, index, overview, blog summary |
Context matters: A source is not inherently primary or secondary. The classification depends on your question. For a claim about a historical event, a diary is primary. For a claim about historical interpretation, that same diary is secondary evidence. For a claim about the state of historiography, a review article is primary.
A 2023 newspaper article reports on a 2022 scientific study. For the claim “the study found a correlation,” is the newspaper article primary or secondary? For the claim “the newspaper reported on the study,” is it primary or secondary?
Show feedback
Answer: For the claim about the study’s findings, the newspaper is secondary (it reports on the original study). For the claim about what the newspaper reported, the newspaper is primary (it is the direct source of that information). The classification shifts with the claim.
6.5 Relevance, Scope, and Context
A source may be credible but irrelevant to the precise claim. Relevance concerns the alignment of the source with the claim’s topic, population, jurisdiction, system, version, language, time, method, outcome, setting, and purpose.
Evaluate alignment involving:
- Topic: Does the source address the same subject?
- Population: Does it cover the same age group, species, or demographic?
- Jurisdiction: Does it apply to the same country, province, or legal system?
- System: Does it concern the same software, hardware, or organisation?
- Version: Does it refer to the same edition or release?
- Language: Is the source in the same language, and if translated, is the translation accurate?
- Time: Is the source from the relevant period?
- Method: Does the source use a method that matches the claim (e.g., experimental vs. observational)?
- Outcome: Does it measure the same outcome (e.g., test scores vs. satisfaction)?
- Setting: Is it relevant to the same context (e.g., workplace vs. classroom)?
- Purpose: Was the source created for the same kind of question?
Common mistake: Using a source that is credible and current but about a different population or jurisdiction. A study of US college students does not directly support a claim about Canadian secondary students.
A 2022 study of blended learning in Australian universities is cited to support a claim about blended learning in Canadian high schools. What relevance problems do you see?
Show feedback
Answer:
- Population: Australian university students vs. Canadian high school students (different age, educational level, culture).
- Jurisdiction: Australian educational policy and curriculum differ from Canadian.
- Setting: University vs. secondary school environments differ in structure, maturity, and pedagogical approach.
- Time: 2022 may be recent, but if the policy or technology has changed, the study may be outdated.
The source may be suggestive but not directly supportive of the claim.
6.6 Currency and Version
Currency concerns whether the source is sufficiently current for the claim. Currency matters especially for law, regulation, policy, software, product features, market information, prices, scientific guidance, office holders, schedules, and emerging technology.
An older source may remain appropriate for historical or foundational claims. For example, a 1953 textbook on DNA structure is authoritative for the discovery of the double helix, but it is not current for CRISPR gene editing.
Common mistake: Assuming that a source is authoritative solely because it is recent, or that an older source is automatically invalid. Currency must be assessed relative to the claim.
You are verifying a claim about the current copyright law in Canada. A 2015 legal textbook is available. Is it sufficiently current? Why or why not?
Show feedback
Answer: A 2015 textbook is not sufficiently current for a claim about current copyright law. Copyright law can change via legislation and court decisions. The authoritative source would be the current Copyright Act and recent case law. The textbook may be useful for historical context but not for a claim about the law today.
6.7 Provenance and Transformations
Provenance concerns the origin and history of information. Students should ask: Where did the claim originate? Has it been edited, summarised, translated, or transformed? Is the original context available? Has the evidence passed through AI-generated text? Can the chain be reconstructed?
Provenance is critical because information can change as it moves between sources. A statistic that was qualified in the original study may become absolute in a summary; a quotation may be truncated; a finding from one context may be applied to another without justification.
An AI response claims “a 2021 study proved that meditation reduces anxiety.” You find the original study and it says “meditation was associated with a small reduction in self-reported anxiety in a sample of 30 university students.” What provenance issues do you see?
Show feedback
Answer:
- The word “proved” is far stronger than “associated with a small reduction.”
- The population (30 university students) is small and specific; the claim generalises to all.
- The outcome (self-reported anxiety) is not the same as clinically diagnosed anxiety.
- The original context included qualifications and limitations that were omitted.
The claim lost critical context and nuance as it moved from the original study to the AI output.
7. CLAIM Walkthrough: First Elements
- Definition: State the exact claim as a proposition. Decompose compound claims into atomic units.
- Guiding questions:
- What is the precise statement being made?
- Is this one claim or multiple claims?
- What are the key terms that need to be verified?
- Example: “Blended learning improves student outcomes by 18%.” → Atomic claim: “The average improvement in student outcomes is 18%.”
- Common mistake: Verifying a compound claim as if it were a single proposition, missing that one part is supported and another is not.
- Evidence requirement: The claim requires quantitative evidence from a study measuring student outcomes.
- Definition: Identify the type of evidence that could support or challenge the claim, and the source types most likely to provide it.
- Guiding questions:
- What kind of evidence would be needed to support this claim?
- Who is most likely to have produced that evidence?
- What source type (primary, secondary, tertiary) would be most authoritative?
- Example: For the 18% claim, the ideal source is the original meta-analysis (secondary) or the underlying primary studies.
- Common mistake: Using a tertiary source (e.g., a blog) as the primary evidence when a primary or secondary source is available and more appropriate.
- Escalation: If the original source cannot be located, escalate to a librarian or subject-matter expert before relying on tertiary sources.
- Residual limitation: Even with the right source type, the evidence may be methodologically weak or inapplicable.
8. Verification-Process Walkthrough
This walkthrough demonstrates how to move from a claim to an evidence requirement, to selecting source types, and then to evaluating authority, relevance, and currency. We use the opening scenario claim: “Blended learning improves student outcomes by an average of 18% compared to traditional instruction.”
- Stage 1: Decompose and classify the claim. The claim is quantitative and causal. It involves a comparison (blended vs. traditional), a population (students, unspecified), an outcome (“student outcomes”), and a specific percentage (18%). It is a comparative effectiveness claim.
- Stage 2: Identify the evidence required. To support this claim, we need a source that: compares blended and traditional instruction; measures “student outcomes” (e.g., test scores, retention, completion); reports a quantitative effect size; and is based on a methodologically sound study or synthesis (RCT, meta-analysis, systematic review). The most appropriate source type is a primary source (the original study) or a secondary source (a high-quality meta-analysis or systematic review). A tertiary source (blog, encyclopedia) is insufficient for a claim this specific and consequential.
- Stage 3: Locate candidate sources and evaluate authority. Candidate sources might include: the meta-analysis cited by the AI (“Centre for Educational Innovation, 2022”); the original primary studies included in the meta-analysis; a government report on blended learning outcomes; a blog post summarising the meta-analysis. Authority evaluation: the meta-analysis is high if peer-reviewed and transparent; the primary studies are high for individual effect sizes; the government report is high for policy relevance but may be secondary; the blog has low authority for a quantitative effect-size claim.
- Stage 4: Evaluate relevance. Does the meta-analysis cover the same population (e.g., secondary students, not university)? Does it measure the same outcomes (test scores, not satisfaction)? Does it compare the same interventions (blended vs. traditional, not blended vs. online-only)? Is the context (e.g., Canadian schools) aligned? If not, the claim may be unsupported or require qualification.
- Stage 5: Evaluate currency. Is 2022 recent enough for educational research? (It may be, but check if newer studies have contradicted or updated the findings.) Have there been major changes in blended learning technology or policy since 2022 that would affect relevance?
- Stage 6: Assess limitations and document. Does the source have conflicts of interest? Are the methods transparent? Is the sample size adequate? Were there important limitations reported in the source that are omitted from the claim? Document all limitations and unresolved questions. If the source does not meet the required authority, relevance, and currency standards, the claim must be corrected, qualified, or rejected.
9. Worked Example: Evaluating a Claim About AI Detection
Claim: “The AI detection tool GPTZero has a 99% accuracy rate in identifying AI-generated text.”
Step 1: Decompose the claim
- Atomic claim: “GPTZero has a 99% accuracy rate.”
- Sub-claim 1: “GPTZero is an AI detection tool.” (definitional)
- Sub-claim 2: “The accuracy rate is 99%.” (quantitative)
- Implicit: “This accuracy rate is based on a valid evaluation.” (methodological)
Type: Quantitative claim about a software tool’s performance. Requires evaluation against a gold standard.
Step 2: Identify evidence required
- Official documentation from the tool’s developer (primary source).
- Independent evaluation study (primary or secondary).
- Peer-reviewed research on the tool’s performance (secondary).
Best source type: Primary source (official developer documentation) and independent primary research (third-party evaluation).
Step 3: Locate candidate sources
- Source 1: GPTZero website (official product page).
- Source 2: A peer-reviewed paper in a computer science journal that evaluates GPTZero.
- Source 3: A news article reporting on GPTZero’s accuracy.
- Source 4: A blog post by a user testing GPTZero.
Step 4: Evaluate authority and relevance
- Source 1 (official website): High authority for the tool’s claimed specifications, but it has a conflict of interest (the developer is self-reporting).
- Source 2 (peer-reviewed paper): High authority for independent evaluation, if it is methodologically sound and peer-reviewed.
- Source 3 (news article): Medium-low authority; it reports on other sources. May be secondary or tertiary.
- Source 4 (blog post): Low authority; it is a personal, uncontrolled test.
Relevance: The peer-reviewed paper is most relevant if it evaluates the same version of GPTZero and uses a representative test set. The official website is relevant but self-interested.
Step 5: Evaluate currency
- AI detection tools are updated frequently. A 2023 evaluation may be outdated if the tool has been updated since then.
- The official website should indicate the current version and date of the accuracy claim.
Currency conclusion: If the only evaluation is from 2023 and the tool has been updated, the 99% claim may not be current.
Step 6: Assess limitations and document
- “Accuracy” is a single metric; it may hide differences in false positives vs. false negatives.
- The test set may not be representative of real-world use.
- The claim “99%” may be based on a specific dataset that does not generalise.
- Conflict of interest: the developer has an incentive to report a high number.
Final assessment: The claim is not verified without access to the official documentation and an independent evaluation. The claim should be qualified: “GPTZero’s developer claims 99% accuracy on a specific test set; independent verification is required.”
10. Guided Learning Activities
- Purpose: Distinguish information from evidence.
- Learning outcome: Distinguish information from evidence.
- Estimated time: 15 minutes.
For each item below, classify it as:
- Information (not connected to a claim)
- Evidence for the stated claim
- Potentially relevant evidence
- Irrelevant information
- Insufficient information
Claim: “Online learning reduces student engagement.”
- A study that found students in online courses reported lower satisfaction than students in face-to-face courses.
- A news headline: “Online learning is on the rise.”
- A dataset of student login times in an online course.
- A quote from a student: “I prefer in-person classes.”
- A report showing that engagement scores in online courses have increased over the last three years.
Show activity feedback
- Evidence for the stated claim — it directly supports the claim (lower satisfaction is a form of reduced engagement).
- Irrelevant information — the rise of online learning does not speak to engagement levels.
- Potentially relevant evidence — login times may proxy engagement, but need interpretation and comparison.
- Insufficient information — a single student’s preference does not support a general claim.
- Contradictory evidence — if engagement scores increased, this challenges the claim.
- Purpose: Classify sources contextually.
- Learning outcome: Classify sources as primary, secondary, or tertiary.
- Estimated time: 15 minutes.
For each source, classify it as primary, secondary, or tertiary for the claim given.
- Source: The original 1953 paper by Watson and Crick on DNA structure. Claim: “Watson and Crick proposed the double helix model in 1953.”
- Source: A 2022 textbook chapter describing Watson and Crick’s work. Claim: “Watson and Crick proposed the double helix model in 1953.”
- Source: A 2023 Wikipedia article on DNA structure. Claim: “Wikipedia’s article on DNA structure was last updated in 2023.”
- Source: A 2020 systematic review of blended learning research. Claim: “The systematic review found a positive effect of blended learning.”
Show activity feedback
- Primary — it is the original record of the proposal.
- Secondary — it interprets and summarises the original work.
- Primary — the Wikipedia article itself is the direct evidence for the claim about its update date.
- Primary — the systematic review is the direct source of its own findings; for the claim about what it found, it is primary.
- Purpose: Evaluate source authority and relevance.
- Learning outcome: Evaluate source authority and relevance.
- Estimated time: 20 minutes.
Claim: “A new drug, Zolpidem, is effective for treating insomnia in older adults.”
Candidate source: A 2021 study published in the Journal of Geriatric Medicine that tested Zolpidem in adults aged 65+ and found a 20% improvement in sleep quality compared to placebo.
Evaluate the source on:
- Author/institution
- Mandate
- Expertise
- Method
- Currentness
- Jurisdiction
- Population
- Scope
- Conflict of interest
- Limitations
Show activity feedback
- Author/institution: Likely academic researchers with geriatric expertise (high credibility).
- Mandate: The journal specialises in geriatric medicine, appropriate.
- Expertise: Authors should have relevant training (verify credentials).
- Method: If it is a randomised controlled trial (RCT), it is strong. If it is observational, weaker.
- Currentness: 2021 is relatively recent for a drug study (check if newer evidence exists).
- Jurisdiction: The study may be from a specific country; drug approvals vary by jurisdiction.
- Population: The study focuses on 65+ (matches the claim).
- Scope: It measures sleep quality (matches “effective”).
- Conflict of interest: Check if authors received funding from the drug manufacturer.
- Limitations: Short-term, small sample, specific population, etc.
Conclusion: The source appears strong but requires deeper verification of conflicts and methodological details.
- Purpose: Select the strongest source type for a claim.
- Learning outcome: Select the strongest source type for a claim.
- Estimated time: 15 minutes.
For each claim, select the most appropriate source type from: primary, secondary, tertiary, or a combination.
- Claim: “The minimum wage in Ontario increased to $16.55 per hour on October 1, 2023.”
- Claim: “AI detection tools have a high false-positive rate for non-native English writing.”
- Claim: “Martin Luther King Jr. delivered the ‘I Have a Dream’ speech on August 28, 1963.”
- Claim: “The Python programming language is used in data science.”
Show activity feedback
- Primary — official government gazette or regulation (current legal source).
- Primary + Secondary — primary: original research studies on AI detection; secondary: systematic reviews or meta-analyses that synthesise multiple studies.
- Primary — the original audio/video recording or official transcript is primary; a news article from the time is secondary.
- Primary — official Python documentation or a reputable survey (e.g., Stack Overflow) that measures language use. Secondary sources (textbooks) may also be authoritative.
11. Exercises
Complete the following exercises. For each, show your reasoning and final answer, then check against the feedback.
- Difficulty: Introductory
- Estimated time: ~10 min
- Learning outcome: Distinguish information, evidence, source, and citation.
For the following scenario, identify: (a) the information, (b) the evidence, (c) the source, and (d) the citation.
Scenario: An AI response states: “According to a 2022 study by Dr. A. Patel in the Journal of Educational Psychology, students who use generative AI for brainstorming show a 15% increase in creativity scores (Patel, 2022, p. 45).”
Show answer
- Information: “15% increase in creativity scores.”
- Evidence: The information becomes evidence for the claim “AI improves creativity” if the study actually found that.
- Source: The study by Dr. A. Patel in the Journal of Educational Psychology (if it exists).
- Citation: “(Patel, 2022, p. 45)” — the reference to the source.
- Difficulty: Introductory
- Estimated time: ~10 min
- Learning outcome: Classify primary, secondary, and tertiary sources.
Classify each of the following as primary, secondary, or tertiary for the claim shown.
- Source: The original journal article reporting a clinical trial. Claim: “The trial showed a 12% reduction in symptoms.”
- Source: A news article summarising the trial. Claim: “The news article reported a 12% reduction.”
- Source: A medical encyclopedia entry on the condition. Claim: “The encyclopedia describes the condition’s symptoms.”
- Source: A meta-analysis of multiple clinical trials. Claim: “The meta-analysis found an overall effect size of 0.8.”
Show answer
- Primary — it is the original record of the trial’s findings.
- Primary — the news article itself is the direct source for the claim about what it reported.
- Tertiary — the encyclopedia summarises primary and secondary sources.
- Primary — the meta-analysis is the direct source of its own finding.
- Difficulty: Intermediate
- Estimated time: ~12 min
- Learning outcome: Evaluate source authority.
Claim: “Vaccines cause autism.”
Source: A blog post from an anti-vaccine advocacy group citing a retracted 1998 study.
Evaluate the source’s authority on the following dimensions: expertise, mandate, methodological quality, review process, conflict of interest, and disciplinary acceptance. Explain your conclusion.
Show answer
- Expertise: The advocacy group likely lacks medical or epidemiological expertise.
- Mandate: The group’s mandate is advocacy, not independent research.
- Methodological quality: The cited study has been retracted due to serious methodological flaws and fraud.
- Review process: No peer review; the blog is self-published.
- Conflict of interest: The group has a strong advocacy interest in promoting a specific viewpoint.
- Disciplinary acceptance: The scientific consensus rejects a causal link; the retracted study is not accepted.
Conclusion: The source has no authority for this claim. The claim should be rejected.
- Difficulty: Intermediate
- Estimated time: ~12 min
- Learning outcome: Evaluate source relevance.
Claim: “The Canadian economy grew by 2.5% in the first quarter of 2024.”
Source: A 2023 report from the World Bank on global economic growth.
Evaluate the source’s relevance. Does it address the claim? Why or why not?
Show answer
Relevance evaluation:
- Topic: The World Bank report may cover global growth, but not specifically Canada.
- Jurisdiction: A global report may not have detailed Canadian data.
- Time: A 2023 report does not cover the first quarter of 2024 (currency problem).
- Population: Not applicable; this is an economic indicator.
Conclusion: The source is not relevant to the specific claim about Canadian growth in Q1 2024. A primary source (Statistics Canada) or a current government report is required.
- Difficulty: Introductory
- Estimated time: ~10 min
- Learning outcome: Evaluate source currency and version.
Claim: “The current version of Python is 3.12.”
Source: A Python textbook published in 2022.
Is this source sufficiently current for the claim? Why or why not? What source would be more appropriate?
Show answer
Currency evaluation: The 2022 textbook is not sufficiently current for a claim about the current version of Python. Software versions change frequently. A textbook is a static snapshot; it cannot report current version.
More appropriate source: The official Python website (python.org) or the Python documentation, which provide current version information.
- Difficulty: Intermediate
- Estimated time: ~10 min
- Learning outcome: Select an appropriate source type for a claim.
For the claim “Canada’s official languages are English and French,” select the most appropriate source type and justify your choice.
Show answer
Most appropriate source: Primary — the Official Languages Act (federal legislation) or the Canadian Charter of Rights and Freedoms. These are the authoritative legal sources.
Justification: This is a legal-constitutional claim. The primary source is the law itself. Secondary sources (legal commentary) can be used for interpretation, but the primary legal text is the authoritative source. Tertiary sources (encyclopedias) are insufficient for a precise legal claim.
12. Homework Problems
Explain why the statement “The sky is blue” is information, not evidence, unless it is used in support of a claim. Give an example of a claim for which it would be evidence.
Show solution design
- Define information and evidence.
- Explain that “the sky is blue” is a statement of fact (information).
- Give an example claim: “The sky appears blue due to Rayleigh scattering.” In this case, “the sky is blue” is evidence supporting the claim (it is the phenomenon being explained).
- Alternatively, for the claim “The sky is blue,” the statement is the claim itself, not evidence.
An AI response includes “(Johnson, 2020)” as a citation. List five things you must verify beyond the fact that the citation exists.
Show answer
- Does the source (Johnson, 2020) actually exist?
- Is the metadata (author, year, title, etc.) correct?
- Does the source support the claim made by the AI?
- Is the source authoritative for this specific claim?
- Is the source current and relevant to the context?
You are verifying a claim about the history of the Canadian Charter of Rights and Freedoms. Classify each source as primary, secondary, or tertiary for the claim “The Charter was enacted in 1982.”
- The original Constitution Act, 1982.
- A 2023 textbook chapter on Canadian constitutional law.
- A Wikipedia article on the Charter.
- A 1983 newspaper article reporting on the enactment.
- A scholarly article analysing the Charter’s impact on Indigenous rights.
Show answer
- Primary — the official legal text.
- Secondary — it interprets and summarises primary sources.
- Tertiary — it organises and summarises primary and secondary sources.
- Primary — it is a contemporary record of the event.
- Secondary — it analyses and interprets the Charter.
Claim: “The COVID-19 vaccine is safe for pregnant individuals.”
Source: A statement from the World Health Organization (WHO) dated January 2024, citing multiple studies and clinical trials.
Evaluate the authority of this source. Consider expertise, mandate, methodological quality, review process, conflict of interest, and disciplinary acceptance. Justify whether this source is authoritative for the claim.
Show solution design
- Expertise: The WHO has global health expertise and access to a wide range of evidence.
- Mandate: The WHO’s mandate includes providing evidence-based health guidance.
- Methodological quality: The WHO synthesises evidence from multiple high-quality studies.
- Review process: WHO guidance is developed through expert review and rigorous processes.
- Conflict of interest: The WHO may have political pressures, but its scientific processes are generally transparent.
- Disciplinary acceptance: WHO guidance is widely accepted in the medical community.
Conclusion: The WHO statement is highly authoritative for this claim, provided it is the most current guidance and cites the underlying evidence.
Claim: “A 2021 study showed that mindfulness meditation reduces workplace stress in tech employees.”
Source: A 2020 study on mindfulness meditation and stress in healthcare workers.
Is the source relevant to the claim? Identify all relevance mismatches.
Show answer
Relevance mismatches:
- Population: Tech employees vs. healthcare workers (different occupations, stress sources, and work environments).
- Time: 2020 study vs. 2021 claim; the study predates the claim.
- Intervention: The 2020 study may have used a different mindfulness protocol than the one implied in the claim.
- Outcome: The measure of “stress” may differ between the studies.
Conclusion: The source is not directly relevant to the claim. It may be suggestive but cannot support the claim without additional evidence linking it to tech employees.
Claim: “The current CEO of Apple is Tim Cook.”
Source: A 2015 biography of Steve Jobs.
Is this source sufficiently current? What source would be more appropriate?
Show answer
Currency evaluation: A 2015 biography of Steve Jobs is not sufficiently current for a claim about the current CEO. Tim Cook became CEO in 2011, but a 2015 biography would not report on the current situation in 2025.
More appropriate source: Apple’s official website (corporate leadership page) or a current financial report.
Claim: “The legal age of consent in Canada is 16.”
Source: A 2022 legal textbook on US criminal law.
What is the jurisdiction problem? What source type would be appropriate for this claim?
Show answer
Jurisdiction problem: The source discusses US law, not Canadian law. The age of consent varies by country and often by province/state.
Appropriate source type: Primary — the Criminal Code of Canada (federal law) or current legal information from the Government of Canada.
An AI response claims: “A 2021 study showed that 90% of students prefer online learning.” You find the original study and it says: “In a survey of 150 students, 90% reported being satisfied with online learning.” What provenance issues do you see?
Show answer
- Sample size: The original study had 150 students; the claim generalises to “students” without qualification.
- Measure: The original measured “satisfaction,” not “preference.” The claim changes the construct.
- Context: The original context (which course, time period, region) is lost.
- Qualification: The original may have included limitations that are omitted.
Conclusion: The claim has been transformed: population generalised, construct changed, context lost, qualifications omitted.
Claim: “The City of Vancouver banned single-use plastics in 2022.”
What source type(s) would you use to verify this claim? Identify the most authoritative source.
Show answer
Most authoritative source: The official City of Vancouver bylaw or council resolution that enacted the ban (primary source).
Other sources: City news releases (primary), news articles (secondary), government websites (primary if official).
Less authoritative: Wikipedia (tertiary), blog posts (tertiary), social media (unreliable).
Think of a claim from your field of study or professional practice that you have encountered recently. Identify the claim, the source(s) that could verify it, and evaluate the authority, relevance, and currency of the most likely source. Write a brief assessment.
Show solution design
- Identify a specific claim relevant to your field.
- State the claim clearly.
- Identify the type of evidence required (e.g., statistical, legal, scientific).
- Identify the most appropriate source type (primary, secondary, tertiary).
- Evaluate authority, relevance, and currency of a specific candidate source.
- Conclude whether the claim is supported or requires qualification.
Example: (Student’s own work will vary.)
13. Mini-Project: Source Authority and Relevance Profile
- Purpose: Apply source-evaluation concepts to a real or supplied claim.
- Learning outcomes: Evaluate source authority, relevance, currency, and provenance; select the strongest source type; justify your selection.
- Estimated effort: 90–120 minutes.
Available claim sets
Choose one claim from the supplied list below, or use a claim from your field (with instructor approval).
- Claim A: “The use of ChatGPT in higher education is associated with a 15% increase in student engagement.”
- Claim B: “The Canada Health Act (1984) requires that all medically necessary services be covered by provincial health insurance plans.”
- Claim C: “The 2023 IPCC report states that global warming is accelerating at a rate of 0.2°C per decade.”
- Claim D: “The current recommended daily allowance of vitamin D for adults is 600 IU.”
Supplied source set (optional)
If you choose Claim A, you may use the following supplied sources. (If you choose another claim, locate your own sources or ask your instructor for a supplied set.)
- Source 1: A 2023 peer-reviewed study in the Journal of Educational Technology titled “ChatGPT and Student Engagement: A Quantitative Study,” by Dr. L. Chen and Dr. M. Okafor. The study found a 15% increase in engagement among a sample of 200 university students in California.
- Source 2: A 2024 blog post on the website “EdTech Innovations” summarising Chen and Okafor’s study, with the headline “ChatGPT Boosts Engagement by 15%.”
- Source 3: A 2025 industry report from a consulting firm, “AI in Education 2025,” which cites Chen and Okafor and states that “engagement increases are likely more modest than initially reported.”
- Source 4: A 2022 textbook chapter on educational technology that mentions the potential of AI but does not include Chen and Okafor.
Required deliverable
Complete the following sections for your chosen claim.
- Claim: State the exact claim you are evaluating.
- Claim type: Is it quantitative, legal, policy, scientific, historical, or professional?
- Evidence required: What kind of evidence would support or challenge this claim?
- Source 1 profile: Describe Source 1 (author, title, date, publication, type).
- Source 2 profile: Describe Source 2.
- Source 3 profile: Describe Source 3.
- Source types: Classify each source as primary, secondary, or tertiary for this claim.
- Authority: Evaluate each source’s authority relative to the claim.
- Relevance: Evaluate each source’s relevance (topic, population, jurisdiction, time, method, outcome).
- Currency: Evaluate each source’s currency for the claim.
- Jurisdiction: Identify any jurisdiction or scope issues.
- Population or scope: Does the source cover the same population as the claim?
- Provenance: Trace the origin and transformations of the claim across sources.
- Limitations: Identify limitations of each source.
- Conflict-of-interest questions: Are there any conflicts of interest?
- Preferred source: Which source is the strongest for this claim? Justify.
- Sources requiring exclusion: Which sources should be excluded? Justify.
- Remaining evidence gap: What evidence is still missing?
- Preliminary conclusion: Is the claim supported, partially supported, or unsupported? State your preliminary conclusion.
Assessment criteria
- Accurate classification of source types
- Rigorous evaluation of authority
- Thorough evaluation of relevance
- Appropriate evaluation of currency
- Clear identification of provenance and limitations
- Justified selection of preferred source
- Clear, evidence-based preliminary conclusion
Show staged hints
Hint 1: Start with the claim. Write the claim as a single proposition. Identify its key terms and concepts. What must be true for the claim to be supported?
Hint 2: Match source type to claim. For a quantitative claim, the most authoritative source is usually the original study (primary) or a high-quality systematic review (secondary). For a legal claim, the primary legal text is authoritative. For a policy claim, the official policy document is primary.
Hint 3: Assess independence. Consider whether the sources are independent or whether they all trace back to the same original report. Source 2 (blog) and Source 3 (industry report) may both derive from Source 1 (the study).
Show illustrative partial example (different claim)
Claim: “The 2020 Canadian federal budget allocated $100 billion to health care.”
Source 1 profile: Government of Canada, Budget 2020: A Plan for a Stronger Canada, official government publication, tabled in Parliament.
- Source type: Primary (official government document).
- Authority: High — the government is the official source of its own budget.
- Relevance: Directly relevant — it is the exact document.
- Currency: 2020 is the budget year; appropriate for the claim.
- Limitations: May include actual vs. projected spending; need to verify the $100 billion figure is for health care specifically.
Source 2 profile: A 2022 news article from the Globe and Mail titled “How the 2020 Budget Shaped Health Care Funding.”
- Source type: Secondary (news reporting).
- Authority: Medium — journalists are not the original source; they report on the budget.
- Relevance: Relevant if it accurately reports the budget.
- Currency: 2022 is later than the budget, but the claim is about the 2020 budget; the source is reporting on something from 2020.
- Limitations: Journalists may summarise, simplify, or misinterpret.
Preferred source: Source 1 (the official budget document) is the most authoritative and relevant. Source 2 can be used to confirm but should not replace Source 1.
14. Self-Check Quiz
Question 1: Which of the following is the best definition of evidence?
Show answer and explanation
Answer: b.
Evidence is always tied to a claim. Information becomes evidence only in relation to a claim and context.
Question 2: A citation does not prove which of the following?
Show answer and explanation
Answer: d.
A citation is a reference, not proof. You must independently verify existence, support, and authority.
Question 3: For the claim “The original research article reported a 20% reduction in symptoms,” which source type is the original research article?
Show answer and explanation
Answer: a.
The original research article is the direct source of the reported findings, making it primary for that claim.
Question 4: Which of the following is a tertiary source?
Show answer and explanation
Answer: c.
A tertiary source summarises primary and secondary sources. Encyclopedias, general reference works, and introductory overviews are tertiary.
Question 5: Source authority is:
Show answer and explanation
Answer: b.
A source may be authoritative for one claim and not for another. Authority is assessed relative to the claim, expertise, and context.
Question 6: A 2015 textbook on software development is cited to support a claim about the current version of a programming language. What is the main problem?
Show answer and explanation
Answer: a.
Software versions change rapidly; a 2015 textbook cannot support a claim about the current version. The official documentation is more appropriate.
Question 7: What is provenance in the context of source evaluation?
Show answer and explanation
Answer: b.
Provenance concerns where the information came from and how it has been edited, summarised, or transformed as it moved between sources.
Question 8: A blog post summarises a peer-reviewed study. For the claim “the study found a positive effect,” the blog post is:
Show answer and explanation
Answer: b.
The blog post interprets and summarises the primary study. It is secondary for the claim about what the study found. If the claim were about what the blog post said, it would be primary.
15. Assignment Connections
In Assignment 5, you will verify an AI-generated passage, evaluate evidence quality, identify errors and omissions, and produce corrected wording. The concepts from this lesson form the foundation of that assignment.
As you complete this lesson, retain:
- Your understanding of the distinction between information and evidence.
- Your ability to classify sources as primary, secondary, and tertiary.
- Your evaluations of authority, relevance, currency, and provenance.
- Your practice in selecting the most appropriate source type for a claim.
- Your documentation of source limitations and evidence gaps.
In Assignment 5, you will apply these skills to every claim in a passage. You will be expected to identify the claim and evidence required; select and evaluate sources; classify each source type; assess authority, relevance, and currency; and document limitations and remaining uncertainty.
Reflect on the following questions in your Assignment 1 journal.
- How does verification differ from asking an AI system to “check itself”? What did you learn in this lesson about why self-checking is insufficient?
- How does information differ from evidence? Provide an example from your own experience or field of study.
- How does source authority depend on the claim? Give an example where a source is authoritative for one claim but not for another.
- Why does source existence not prove claim support? What else must you verify?
- How has your understanding of source evaluation changed after completing this lesson? What will you do differently when evaluating sources in the future?
Note: Your instructor may provide additional guidance on journal expectations. Write in your own voice; there are no model responses.
Show reflection planning checklist
- I have distinguished information from evidence in my response.
- I have explained why source authority is contextual.
- I have connected my reflection to a specific claim or example.
- I have identified a change in my own source-evaluation practice.
- I have used clear, specific language.
16. Key Takeaways
- Information is not automatically evidence; it becomes evidence only in relation to a claim.
- A citation is not proof; it points to a source. You must verify existence, authenticity, and support.
- Primary sources provide direct evidence; secondary sources interpret; tertiary sources summarise. The classification depends on the claim.
- Authority is contextual — it depends on the relationship between the source and the claim.
- Relevance requires alignment of topic, population, jurisdiction, time, method, and context.
- Currency is not absolute; it depends on the claim. Some claims require very recent sources, while others do not.
- Provenance — the origin and transformation history of information — is critical for detecting misrepresentation.
- Never rely on popularity, search ranking, or professional appearance as proxies for authority or accuracy.
17. Key Terms
- Information
- Data, statements, observations, records, documents, or communications that may or may not be evidence.
- Evidence
- Information used to support, qualify, challenge, or evaluate a particular claim.
- Source
- The person, record, dataset, document, publication, institution, or other origin from which information is obtained.
- Citation
- A reference that identifies or points to a source. It does not prove existence, support, or authority.
- Primary source
- A source that provides direct, original, or first-hand evidence relevant to the field or question.
- Secondary source
- A source that interprets, analyses, critiques, compares, or synthesises primary material.
- Tertiary source
- A source that summarises or organises information drawn from primary and secondary sources.
- Authority
- The quality of a source based on its expertise, mandate, methodological quality, transparency, and accountability relative to a claim.
- Relevance
- The alignment of a source with a claim’s topic, population, jurisdiction, system, version, time, method, outcome, and purpose.
- Currency
- The timeliness of a source relative to the claim. Some claims require very recent sources; others do not.
- Provenance
- The origin and history of information, including any transformations, summaries, translations, or omissions.
- Contextual authority
- The principle that a source’s authority depends on the claim and the context, not on an absolute status.
18. Looking Ahead: Lesson 11.2
You have now learned to distinguish information from evidence, sources from citations, and to evaluate authority, relevance, currency, and provenance. These skills are the foundation of responsible verification.
In Lesson 11.2, you will build on this foundation to:
- Align evidence with specific claims.
- Distinguish evidence that supports, partially supports, qualifies, contradicts, or does not address a claim.
- Evaluate source independence and distinguish corroboration from repetition.
- Apply triangulation cautiously and preserve disagreement and uncertainty.
These skills will allow you to move from finding a source to interpreting what the source actually means for the claim.
19. Sources & Further Reading
- Primary textbook: Chapter 8: Verification, Evidence, and Epistemic Responsibility (COMP 3270 course text). [PAGE NUMBER TO BE CONFIRMED]
- Supporting: Appendix C: Verification Toolkit (COMP 3270).
- Supporting: Appendix A: Generative AI Glossary (COMP 3270).
- Council of Science Editors. (2019). Scientific Style and Format. (For citation guidance — check institutional requirements.)
- Purdue Online Writing Lab (OWL). “Evaluating Sources: Where to Begin.”
- University of Toronto Libraries. “Evaluating Sources.”
- Stanford History Education Group. “Lateral Reading.”
Additional authoritative sources may be provided by your instructor or librarian. If you are unable to locate a source, use the supplied-source alternatives in the activities.