IB Digital Society Paper 2 Drilling: Four Source-Based Questions (2/4/6/12) with Model Answers

Question 1

2 marks = reason 1 + evidence 1
(1) Reason (1 mark): to stop cyberbullying conflicts carrying over into lessons (or: so students concentrate on learning at school). (2) Evidence (1 mark): of the 17 cyberbullying cases the counselling team handled this year, 11 involved conflicts that began on social platforms after school and continued into lessons.
Where marks go: copying the whole letter without separating reason from evidence, or reporting the parents’ view as the principal’s (Source A carries only the principal’s voice).

Question 2

(a) 2 marks: the indicator turns effect into a before/after comparison
Source B defines the effect as the change in the self-reported average for the same respondents: 4.2 hours before the policy, 3.1 hours after it, a fall of 1.1 hours (about 26.2%), and then treats that fall as evidence that the ban worked.
(b) 2 marks: one limitation, explained
(1) An average hides the distribution: a few heavy users can dominate the mean while most students do not change at all. (2) Self-report bias: respondents under-report their own use, and after the policy they know what answer is expected. (3) The indicator does not match the aim: screen time is not the same as distraction or falling attainment, and using less is not the same as learning to manage it.

Question 3

6 marks = three comparative points, 2 marks each (each with evidence from the sources)
(1) Different starting points: Source B centres on measured behaviour (usage and support) and treats the policy as a measurable intervention; Source C centres on rights and learning tools and treats the policy as a limit on student autonomy. (2) Opposite readings of the same fall in use: Source B takes the drop as success, while Source C argues a ban only pushes use ‘out of sight’; the same phenomenon is evaluated in opposite ways. (3) Different kinds of evidence: Source B is a quantitative survey of 500 respondents (percentages), Source C is a positional statement of principle and experience with no data. Similarity (a fourth point to mention; the six marks still count three comparative points): both accept that phones are embedded in school life and both argue from students’ interests.
Where marks go: describing the two sources side by side without comparing them – no ‘whereas’, ‘by contrast’ or ‘both’.

Question 4

12 marks = level descriptors (0 / 1-3 / 4-6 / 7-9 / 10-12)
For 10-12 an answer must: (1) define ‘complete ban’ – scope (all day on campus versus lessons only), who it binds and what exceptions exist (medical or special educational needs); (2) argue for it with the sources: 11 of the 17 cases in Source A carried over into lessons, so conflict follows students onto campus, and the 1.1-hour (26.2%) fall in Source B can stand as evidence of reduced opportunity; (3) argue against it with the sources: Source C shows a ban pushes use out of sight and removes a learning tool, and the average in Source B hides the distribution; (4) use course knowledge: the concepts of power, values and ethics and systems explain how a school rule allocates power, and change explains behaviour moving rather than disappearing; (5) state explicit criteria: proportionality (is the restriction the least necessary), enforceability (can it be applied fairly, and to whom), effects on different groups (special educational needs, household income) and whether a less intrusive alternative exists (no use in lessons plus a designated area); (6) conclude on how far the claim holds and offer a defensible middle course – no phones in lessons and examinations, a designated area at break, sanctions for misuse rather than for possession.
Where marks go: listing both sides without judging; giving opinion with no source used at all; reading the percentage in Source B as proof that the ban worked.

Question 1

2 marks = claim 1 + figure 1
(1) Claim (1 mark): automation speeds up processing and cuts errors (either one). (2) Figure (1 mark): average processing time fell from 14 days to 3; or 780 of 1200 cases (65.0%) were decided by the system alone.
Where marks go: ‘faster and better’ with no figure quoted, or presenting the audit’s figures as the department’s claim (the two sources have different origins and purposes).

Question 2

(a) 2 marks: how it is used
Source B measures how far the process can be understood by the share of decisions carrying written reasons (507 of 780, or 65.0%), and sets that against the unpublished criteria: reasons exist, but the basis of the decision cannot be tested.
(b) 2 marks: why that is not enough – pick one and explain
(1) A reason is not a criterion: written reasons usually restate the outcome, so an applicant cannot see which piece of data or which rule produced the refusal, and cannot point to an error. (2) Untestable means unappealable: with the criteria unpublished and no version history, an applicant cannot show that the system used outdated or wrong data. (3) The review rate is itself a threshold: only 96 cases (8.0%) sought review, and the most vulnerable find it hardest, so a low rate is not proof that the decisions were right.

Question 3

6 marks = three comparative points, 2 marks each (each with evidence from the sources)
(1) Different objects of judgement: Source A judges efficiency (processing time, error rate), Source C judges protection of rights (being informed, review, appeal). (2) Opposite readings of the same fact: Source A presents a majority of automated decisions as progress, while Source C treats the same situation (applicants not knowing why they were refused) as the risk – the two draw opposite conclusions about more automation. (3) Different demands on disclosure: Source A says nothing about publishing the criteria; Source C explicitly demands published criteria, human review and an independent appeal. Similarity (a fourth point to mention; the six marks count three comparative points): both argue from the interests of citizens and applicants, and neither asks for the system to be abandoned.
Where marks go: describing what each source says without naming the nature of the difference (efficiency versus rights).

Question 4

12 marks = level descriptors (0 / 1-3 / 4-6 / 7-9 / 10-12)
For 10-12 an answer must: (1) define the terms – a decision affecting rights covers refusal, withdrawal of a qualification or suspension of a benefit, and ‘independent’ means a body outside the deciding department; (2) argue for it with the sources: Source B shows 65.0% of decisions carried reasons while the criteria stayed unpublished, and 25.0% of reviews were overturned, so errors do occur and the present disclosure is not enough for applicants to find them, while Source C shows the most vulnerable find appeal hardest; (3) argue against it with the sources: the efficiency and error-rate gains in Source A are a public good, fully published criteria can be gamed (applicants adjusting to the rule), disclosure may touch personal data or enforcement-sensitive material, and universal independent appeal may push processing times back up for simple cases; (4) use course knowledge: power, values and ethics and systems explain how automation concentrates judgement in the designer, and change explains how institutions adapt to technology; (5) state criteria: proportionality (is the limit on disclosure the least necessary), accountability (who can be held responsible), feasibility (are there resources) and consistency (are like cases treated alike); (6) conclude on the conditions – for example publishing a summary of the criteria with a version history, and scaling the depth of appeal to the impact of the decision (high-impact decisions need human review and independent appeal), noting that disclosure and efficiency are not necessarily in opposition.
Where marks go: answering ‘it should be published because it is fair’ without dealing with cost and gaming; never citing the overturned share in Source B; treating ‘independent’ as review by a higher unit of the same department.