# Data and data analysis: foundations, security and dilemmas

3.1A–I · supplied chapter pp. 57–80 · four teaching arcs; choose to fit prior learning

Chapter reference edition · 2026-10-11-v6

## 1. Data records part of a situation

A viewer might watch a video because they enjoy it, need it for a class or disagree with it. The recorded action is not identical to the viewer’s reason. Data analytics uses observations to find patterns or make predictions. We begin by distinguishing a recorded value from an inference. This matters when a platform decides what to show next. Our examples use invented records. We will not collect personal histories or require students to enter accounts to explore the mechanism.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 72–73.

## 2. One reading becomes four different claims

Purpose: Context and judgement add meaning; more data alone is not wisdom. Say: A number without a unit, place or time is hard to interpret. Information adds those relationships. Knowledge draws a usable conclusion; wisdom weighs a course of action and its limits. A sensor cannot settle the safety decision by itself. Ask: In 2 minutes, distinguish data from information using a new rainfall example. Then explain what extra reasoning a decision requires. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 57–58. Original classroom examples are fictional. 

## 3. The question determines which data you need

Purpose: Quantitative and qualitative data describe different aspects of a situation. Say: Financial, medical, meteorological, geographical and scientific are context categories. Quantitative values support numerical comparisons; qualitative descriptions convey experiences and meanings. Scientific data can be unpublished observations as well as published research; publication is not what makes a measurement data. Ask: Name a qualitative and a quantitative record for a fictional shop. Explain why neither alone establishes customer satisfaction. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 59–61. Original classroom examples are fictional. 

## 4. A photo has content and metadata

Purpose: Metadata can reveal something the visible image does not. Say: The pixels are the visible content. Descriptive fields help systems organize and interpret a file. Removing a visible person does not necessarily remove location or date records. Metadata can be useful and sensitive at the same time. Ask: A student removes their face from an image but leaves location metadata. Explain one remaining privacy risk. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 62. Original classroom examples are fictional. 

## 5. A sensor measurement becomes a digital record

Purpose: Sampling and conversion create a representation of a physical quantity. Say: An analogue-to-digital converter expresses sampled signal values as discrete digital values. The chosen resolution, timing and calibration constrain the measurement. It records part of the world, not the whole phenomenon. Ask: Explain why the stored number is a representation of temperature. Name one way poor sampling or calibration could limit it. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 61. Original classroom examples are fictional. 

## 6. Find a pattern, or link records about an entity?

Purpose: Data mining and data matching answer different questions. Say: Matching identifies which records refer to the same entity. Mining investigates patterns or anomalies within records. The operations can be combined, but a mistaken match can contaminate the pattern. Association does not establish cause. Ask: A shop joins returns to purchases and then looks for repeated return patterns. Name each operation and one limitation. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 62–63. Original classroom examples are fictional. 

## 7. Check A–C: explain the meaning of a record

Purpose: Use terminology to justify a distinction, then correct one weak inference. Say: Listen for contextual meaning rather than merely “processed”. Accept a defensible library example: loan ID matches returns; repeated late-return patterns are mined. A loan record alone cannot prove enjoyment. Ask students to correct a specific weak term and retry on a shop. Ask: Close notes. Use a fictional library system to distinguish data/information, content/metadata and matching/mining. Explain one unsupported inference. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 57–63. Original classroom examples are fictional. 

## 8. Five stages manage a record over time

Purpose: Preservation and destruction are distinct decisions. Say: Creation can be manual or automatic. Storage concerns location, permissions and protection. Usage transforms or shares data. Preservation maintains suitable data for a stated future need. Destruction requires a method that fits the media and all relevant copies. Ask: Identify two decisions at different stages that could protect a participant. Explain why simply keeping everything is not a plan. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 63–64. Original classroom examples are fictional. 

## 9. Primary or secondary depends on the investigation

Purpose: The relationship to collection matters, not whether the file is online. Say: Original collection for the current purpose is primary. Reuse of data someone else collected is secondary. Check methods, date, units and population before using it for a changed purpose; online versus paper does not settle the distinction. Ask: A school downloads a dataset originally collected by a weather agency. Explain its status for each organization and one reuse limitation. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 65. Original classroom examples are fictional. 

## 10. An entity becomes records with typed fields

Purpose: A database schema states what each field can hold. Say: Entities are things represented by the system. Attributes are properties of them. Fields store those attributes, with types such as integer, floating point, string, character, Boolean or date. Digits can be identifiers: leading zeros and formatting matter for a phone number. Ask: Choose suitable fields/types for a fictional event booking. Explain why a telephone number should not be treated as a quantity. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 65. Original classroom examples are fictional. 

## 11. Keys link entities without repeating every detail

Purpose: The primary key identifies a record; a foreign key refers to another table. Say: A unique key avoids ambiguity. A foreign key records a relationship. One book can have many loan records; the book title need not be copied into every loan. Wrong or missing links undermine integrity even if every value has a valid type. Ask: Explain the BookID link in your own words. What fails if two books share the same primary key? Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 65–67. Original classroom examples are fictional. 

## 12. Suitable is different from accurately copied

Purpose: A valid value can still be the wrong value. Say: Validation checks permitted form or range: type, length, input mask, range or lookup. Verification compares entry with the intended source or a second entry. Neither guarantees the source itself is true. Explain the exact error each can detect. Ask: A date passes the allowed format but was copied from the wrong row. Which check passed, which is needed, and why? Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 66. Original classroom examples are fictional. 

## 13. Classify risk before deciding access

Purpose: Sensitivity and purpose determine permissions; categories may overlap. Say: Categories guide the design, not just vocabulary. A person handling attendance may need a count without medical details. Even public data needs integrity protection. Check jurisdiction before presenting classification words as legal definitions. Ask: Which event record should volunteers see: attendance count or a health request? Explain a purpose-based access decision. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 66–67. Original classroom examples are fictional. 

## 14. Choose a chart that fits the question

Purpose: A chart needs a suitable comparison, units and honest scale. Say: Tables preserve exact values; charts emphasize a relationship. Bar lengths compare categories, while lines connect ordered time observations. An infographic combines evidence and explanation, but decoration cannot replace labelled axes, units or source dates. Ask: Choose a chart for rainfall totals at three sites and a different chart for one site over seven days. Explain the axes and one misleading choice to avoid. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 67–68. Original classroom examples are fictional. 

## 15. Check D–F: build a defensible data record

Purpose: Explain a lifecycle, a relationship and a representation. Say: Check a key link, a valid-but-wrong example and a chart suited to a question. Correct one defect before asking students to repeat the design in a community event. This is a short formative application, not a programming or database implementation requirement. Ask: Sketch a fictional library record and a linked loan. Explain one validation/verification distinction, one lifecycle decision and one suitable chart. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 63–68. Original classroom examples are fictional. 

## 16. Encryption changes readable data into ciphertext

Purpose: Symmetric encryption uses the same secret key to encrypt and decrypt. Say: Encryption protects confidentiality by transforming data under a key. The authorized party recovers plaintext. It does not make the original record accurate or decide who should be permitted to use it. A Caesar cipher can demonstrate substitution but is not modern security. Ask: If an attacker obtains both ciphertext and the secret key, explain why the protection is weakened. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 68–69. Original classroom examples are fictional. 

## 17. Public-key encryption separates the two key roles

Purpose: Encrypt for the recipient with their public key; decrypt with their private key. Say: The recipient shares their public key while retaining the private key. Do not teach that the sender shares the recipient’s private key. Modern secure connections combine asymmetric key establishment/authentication and symmetric protection; SSL is historical terminology, TLS is the current protocol family. Ask: Whose public key should a sender use for a confidential message to a recipient? Explain who holds the private key and why it stays secret. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 69. Original classroom examples are fictional. Precision update checked 11 Oct 2026: https://csrc.nist.gov/glossary/term/public_key_cryptography

## 18. Masking and encryption solve different problems

Purpose: Replacing visible details does not automatically remove re-identification risk. Say: Masking substitutes or obscures selected values for a purpose such as software testing. Encryption is reversible with the relevant key. Neither by itself proves anonymity. Evaluate what remains and what can be linked; do not promise that name removal guarantees privacy. Ask: A dataset removes names but keeps exact home locations and rare diagnoses. Explain why “anonymous” needs further evidence. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 69. Original classroom examples are fictional. 

## 19. Deletion is not a disposal guarantee

Purpose: Choose sanitization for the storage medium, sensitivity and required assurance. Say: Destruction is a lifecycle decision; sanitization is the method and verification. A recycle bin, a pointer deletion or an ordinary format does not necessarily sanitize all storage. Degaussing concerns suitable magnetic media, not an SSD. Students analyse the decision; they must not erase school devices. Ask: A school is retiring an SSD containing confidential records. Explain why “delete the folder” is insufficient and what an authorized technician must verify. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 70. Original classroom examples are fictional. Current layer: NIST SP 800-88 Rev.2, Sept 2025, checked 11 Oct 2026. https://csrc.nist.gov/pubs/sp/800/88/r2/final

## 20. A linked ledger protects record history, not truth at entry

Purpose: Hash links and network agreement make some changes detectable or difficult. Say: A distributed ledger is replicated across participants. Cryptographic links and consensus rules resist undetected alteration of accepted history. They do not certify a real-world claim before entry. Identity, inspection and governance remain important; personal-data retention can conflict with difficult-to-change records. Ask: A seafood supplier enters a false catch location into a ledger. Explain what a blockchain could protect and what it cannot establish. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 71–72. Original classroom examples are fictional. 

## 21. Check G: match the safeguard to the failure

Purpose: Use a safeguard’s mechanism and limit, not just its name. Say: Require the correct key roles and a specific limit. Reject “encryption makes data true” and “blockchain means nobody can lie”. If unclear, trace one message from readable to ciphertext to readable; then apply it to a different organization. Ask: For a fictional community centre, explain symmetric/public-key encryption, masking/deletion and one blockchain limitation. Select one suitable safeguard and a residual risk. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 68–72. Original classroom examples are fictional. 

## 22. Big data and data analytics

The textbook characterizes big data using four Vs. Data can be structured in a table or arrive in formats such as video, text and images. Analytics can describe what happened, detect patterns or support a prediction. A classroom table is small data, but it helps us reason about a process that operates on a much larger collection. A numerical output is not automatically a reliable answer. Ask how the records were collected, what they represent and how the analysis matches the question.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 72–73.

## 23. Volume and velocity

Consider a match where sensors generate records throughout play. Volume concerns how many records accumulate. Velocity concerns the rate at which new records arrive and whether analysis can keep up. A system that stores a large archive may have high volume without fast current arrival. A live stream may require quick decisions before every record can be manually checked. Explain which characteristic matters to a particular task instead of treating big data as a single vague advantage.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 73.

## 24. Variety and veracity

A platform may combine a video, its caption, a device setting and an interaction log. These are different kinds of data. That variety creates opportunities, but joining formats accurately can be difficult. Veracity concerns whether the values are trustworthy for the intended use. Repeated records, missing values, outdated labels and false accounts can all distort an analysis. Having many formats does not establish good quality. A large dataset can remain unrepresentative even when its individual records are accurate.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 73.

## 25. Small and clean, or large and messy?

These invented datasets force us to name a purpose. Dataset A may be suitable for checking one class’s activity, but it may not represent the whole school. Dataset B might capture a wider range of people, yet its errors require investigation. We cannot decide which is better without knowing the question, sampling, missingness and quality controls. Data cleaning can improve validity while accidentally removing groups if missing values are unevenly distributed. Make the decision and its assumption visible.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 73–74.

## 26. A missing value is an unresolved question

Suppose an item has no completion value. Replacing it with zero can push the creator down the ranking even though zero completion was never observed. Removing the item avoids the calculation but also removes its opportunity for exposure. Estimating a value may help, yet it relies on an assumption that needs checking. A sensible choice depends on the task and consequence. Record what you chose and how the result might change under another plausible choice.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 73–74.

## 27. An association does not establish a cause

Items with more exposure often receive more interactions. This could reflect relevance, the number of opportunities to be seen or both. Looking only at interaction totals cannot distinguish those explanations. A carefully designed comparison might hold exposure constant or compare a changed allocation, but even then we need to examine other factors and limits. The important habit is to state the measured relationship first, then ask which alternative explanations the evidence has ruled out.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 73.

## 28. The data lifecycle creates decisions at every stage

The chapter places big data inside a lifecycle. At collection, we decide what to record and whom to include. Storage involves access, security and organization. Use involves analysis and sharing for a purpose. Preservation and destruction concern how long records remain and what happens when they are no longer needed. This week’s assigned reading begins the collection dilemmas in section 3.1I. It does not teach all later lifecycle dilemmas. A diagram of a lifecycle is a map, not proof that every stage has been adequately managed.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 72, 74.

## 29. Collection can shape the eventual result

The assigned page raises ethical collection, consent and excessive collection. We can ask what information is actually needed to answer a question, and whether a group is missing before analysis begins. Collecting an additional personal detail can create a privacy cost without improving the decision. Applicable rules depend on the setting and jurisdiction, so this is an ethical design discussion rather than a universal legal checklist. Explain the purpose, the proposed data and the effect of collecting or omitting it.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 74.

## 30. Analytics can support several purposes

The textbook gives historical examples from finance and sports. The enduring idea is the use of data to inform a decision. Describing last month’s activity, flagging an unusual transaction and predicting interest are different tasks. A prediction expresses an estimated outcome and should be evaluated against later observations. The examples do not make analytics infallible. A false alert can inconvenience someone, while a missed alert can leave a problem undetected. Identify the task and the cost of each kind of error.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 73.

## 31. A bounded big-data inquiry

A manageable inquiry identifies a particular use rather than asking whether all big data is good. For a platform, we could investigate whether recorded interactions represent new creators as well as established ones. The real case file gives a provider account and a regulator account with different purposes. Neither offers a complete dataset of individual experiences. Use the source passport to record the locator, claim, evidence and limitation. A limitation helps you decide what a source is useful for rather than simply dismissing it as biased.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 74.

## 32. Reliability and integrity are separate tests

Purpose: Accurate, complete data can become unreliable through unauthorized changes. Say: Update, verify and evaluate sources to improve reliability. Control access and track alterations to protect integrity. Biased sampling, stale records, malware and entry errors can cause different failures. Automatic collection removes some manual errors but introduces others. Ask: A telephone number is outdated; a second record was maliciously changed. Explain each problem and a different corrective action. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 75–76. Original classroom examples are fictional. 

## 33. Privacy and anonymity are different conditions

Purpose: A person can be identifiable while their information remains restricted. Say: Anonymity concerns whether someone can be identified; privacy concerns access, use, surveillance and control. Legitimate advice or expression may need protection. Accountability and safety may create competing demands; judge a specific use, not all anonymity. Ask: Explain how a named record can be private and an unnamed post can reveal a person. Name one protection beyond removing a name. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 76–79. Original classroom examples are fictional. 

## 34. Data protection starts before collection

Purpose: Purpose, minimization and accuracy constrain the data pipeline. Say: The GDPR groups lawfulness, fairness and transparency as one principle; purpose limitation, minimization and accuracy are distinct. Consent is not the only possible lawful basis. Use the framework to analyse the fictional event, not to issue legal advice about school practice. Ask: An event form requests birth date and later sells an advertising list. Identify two relevant principles and justify an improvement. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 77–78. Original classroom examples are fictional. Current official framework checked 11 Oct 2026: https://www.edpb.europa.eu/topics/key-gdpr-concepts/basic-principles_en

## 35. Retention and responsibility finish the protection plan

Purpose: A protection promise needs a retention rule, safeguards and accountable decisions. Say: Access, correction and erasure are rights with specific conditions; they are not an unconditional promise that every record disappears instantly. Reconcile justified retention, backups and responsibilities. Ask for a purpose, a decision owner and an auditable action rather than a vague privacy slogan. Ask: Explain why “keep it forever in case it is useful” needs challenge. Name one record the organization should keep about its own decisions. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 77–79. Original classroom examples are fictional. Current official framework: https://www.edpb.europa.eu/topics/key-gdpr-concepts/basic-principles_en

## 36. A result difference does not establish its cause

Purpose: Compare what the report measured with what the textbook infers. Say: The supplied chapter summarizes demographic effects and attributes them to training images. The primary NIST report explicitly separates measurement from causal explanation. Models, image quality and datasets need investigation. Do not assign a student’s race or collect face images for this exercise. Ask: Identify one supported finding and one further investigation needed before saying training-set diversity caused all error differences. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 75. Original classroom examples are fictional. Primary historical source checked 11 Oct 2026: https://www.nist.gov/news-events/news/2019/12/nist-study-evaluates-effects-race-age-sex-face-recognition-software

## 37. Check I: make a conditional data decision

Purpose: A defensible judgement joins mechanism, evidence, stakeholders and a safeguard. Say: A strong response explains why a particular record and purpose create the consequence. Accept a conditional policy grounded in purpose, access and retention; require a residual risk. For Paper 2-style practice, compare the textbook’s causal assertion with what NIST actually establishes, then give a qualified synthesis. Ask: Close notes. Recommend a data policy for a fictional community event. Explain one lifecycle dilemma, one principle, a stakeholder trade-off and a condition that could change your judgement. Practice feedback: use the worked exhibit to check the distinction; keep the first attempt and append a correction. Secure: apply the distinction to a different venue. Partial: isolate the disputed operation and compare the two cases. Unclear: reconstruct the first worked example before retrying. Later return: next lesson, close notes and explain this distinction in a library or community-centre system. These are formative practice responses, not registered assessment items.

Source: Textbook layer: Hodder 2022, supplied chapter pp. 74–80. Original classroom examples are fictional. 

## 38. Original excerpt check: pp.72–74

Say: The supplied chapter now has an explicit teaching route. Earlier excerpt questions and IDs remain unchanged. Use checks A–C, D–F, G and I for the added sections; the existing Google Data branch samples pp.72–74 only. Expanded checks are formative export/print hand-in to the teacher, not new automatic marks.

Source: Bomfim et al., Digital Society for the IB Diploma, Hodder, 2022. Printed pages cited per slide. Original paraphrase. Textbook connection: pp. 72–74.