CUT SSI21AB: Social Sciences Informatics – Using Technology for Social Research

Social Sciences Informatics examines how digital tools, data infrastructures, and computational methods can strengthen social research—while also raising ethical, legal, and methodological concerns. In a South African context, students must connect informatics skills to real research problems such as inequality, service delivery, migration, youth employment, gender-based violence, education outcomes, and community development. This study guide focuses on what you’re likely to be expected to know for CUT SSI21AB: Social Sciences Informatics – Using Technology for Social Research, with emphasis on practical workflows, critical thinking, and assessment-ready explanations tailored to Central University of Technology (CUT) and the broader Southern African higher education/TVET environment.

1) Foundations of Social Sciences Informatics and Technology-Assisted Research

Social Sciences Informatics is not “just using software.” It is a structured approach to social research where information systems, data technologies, and computational thinking are integrated into research design, fieldwork, data management, analysis, and dissemination. The “informatics” part matters because it forces researchers to treat data as an organized asset—governed, documented, reproducible, and ethically collected—rather than as informal notes that happen to be analysed later.

In an exam or practical assessment context for CUT SSI21AB, you should expect questions that test whether you can (1) define core concepts accurately, (2) describe research workflows step-by-step, and (3) evaluate ethical trade-offs and methodological risks.

1.1 Key Concepts: Data, Information, Knowledge, and Evidence

A solid starting point is understanding how informatics frames knowledge generation.

• Data: Raw facts collected from observations, sensors, surveys, administrative records, interviews, or digital traces.
• Information: Data processed to be meaningful (e.g., coded variables, cleaned datasets, summaries).
• Knowledge: Interpretations that explain patterns (e.g., theories supported by evidence).
• Evidence: Justified claims derived from data, including limitations and uncertainty.

In social research, students often mistakenly treat evidence as “what the data shows” without connecting to theory, sampling, measurement validity, and bias. A key exam-ready argument is: informatics improves the handling of data, but it does not automatically improve the quality of evidence.

1.2 Technology in Social Research: What Changes and What Doesn’t

Technology can transform at least four research dimensions:

1. Access to participants and contexts
   • Online surveys, WhatsApp-based recruitment, remote interviews, and digital ethnography.
2. Measurement and data richness
   • Time stamps, GPS coordinates, call detail records (where legally accessible), and multimedia data.
3. Scalability
   • Larger samples, faster transcription, automated coding support, and dashboarding.
4. Speed of iteration
   • Rapid prototyping of instruments and early data quality checks.

But several fundamentals still remain:

• Research ethics still govern consent, privacy, and harm.
• Sampling and measurement validity remain central.
• Interpretation still requires theoretical reasoning and reflexivity.
• Validity and reliability must be demonstrated, not assumed.

In exam answers, a strong pattern is to present: Technology adds capabilities, yet it introduces risks that must be actively managed.

1.3 Research Lifecycle with Informatics: An Exam-Answer Framework

A reliable approach is to describe the research lifecycle as a sequence of stages, each with specific informatics tasks:

1. Problem formulation and operationalisation
2. Research design and instrument development
3. Data collection (fieldwork/digital capture)
4. Data management (storage, documentation, versioning)
5. Data processing and cleaning
6. Analysis (quantitative, qualitative, mixed methods, computational)
7. Validation, interpretation, and reporting
8. Dissemination and long-term stewardship

For CUT SSI21AB-style assessment, it’s useful to add a “risk lens” to each stage:

• Are there privacy risks?
• Is there measurement error (e.g., biased app responses)?
• Could there be digital exclusion (participants without connectivity)?
• Is data transfer secure?
• Are we able to reproduce the analysis?

1.4 Informatics Tools as Research Infrastructure

In social research technology, tools are best thought of as infrastructure:

• Devices: smartphones, tablets, recording tools, scanners
• Software: survey platforms, spreadsheet and statistical packages, qualitative tools, mapping tools
• Data formats: CSV, JSON, audio formats, transcription files, GIS layers
• Databases and repositories: secure storage, institutional repositories
• Automation: pipelines for cleaning, scripts for reproducible analysis, automated transcription assistance
• Collaboration: version control systems, shared coding frameworks

A recurring exam theme is that a researcher must understand how data flows from participant to server to analysis environment to report. Many marks can be gained by drawing a simple data flow diagram in words.

2) Data Collection, Digitisation, and Fieldwork Technologies in South African Contexts

This section focuses on what you can realistically do with technology for social research and how to justify methodological decisions. It also addresses the South African realities of connectivity, language diversity, ethics boards, and community engagement.

2.1 Designing a Technology-Enabled Data Collection Strategy

A technology-enabled strategy is not automatically “better.” The correct approach is to match technology to research objectives and context. Consider typical social research questions relevant to South Africa:

• How do youth perceptions of job opportunities change over time?
• What barriers do households face when accessing health or social services?
• How do community members experience safety and trust in local policing?
• How are social media narratives influencing collective action?

A technology strategy might combine:

• Online or mobile surveys for scale
• Semi-structured interviews (audio recorded) for depth
• Focus groups with structured facilitation and transcription
• Observational methods with digital note-taking
• Document analysis using digital text extraction from PDFs

In exam answers, you can score well by explicitly linking each method to:

• sampling approach,
• instrument type,
• data format produced,
• analysis plan,
• ethical considerations.

2.2 Survey Technology: From Paper to Mobile and Web-Based Instruments

When using digital surveys, researchers must manage issues like question ordering effects, screen constraints, and usability. Typical design principles:

• Plain language and local terminology
• Progressive disclosure (don’t overload participants)
• Skip logic (reduce missingness)
• Validation checks (range checks for numeric responses)
• Time stamps for fieldwork tracking (with privacy safeguards)

Common risks:

• Digital exclusion: participants without smartphones or data bundles.
• Response bias: people may answer differently when surveyed on a phone vs interviewer-administered.
• Technical failure: connectivity issues causing incomplete responses.

A strong exam-style counter-argument is to note that paper also has problems—manual coding errors and slower data entry—so the key question is which bias is less harmful given the population.

2.3 Interviewing with Digital Recording and Transcription

Digital recording supports accurate capture, but it introduces privacy and data governance challenges.

Good practice workflow for interviews:

1. Explain study purpose using an approved consent script.
2. Obtain informed consent for audio recording (not just for participation).
3. Record using a stable device and label files with participant codes (not names).
4. Transfer recordings to secure storage immediately.
5. Create an audit trail: date/time, interviewer, location type, and any deviations from protocol.
6. Transcribe (manually or with assisted transcription) and check transcripts against recordings.

Quality management steps:

• confirm transcription accuracy for local language phrases,
• standardise spelling for names of organisations and places,
• maintain a coding-ready format (speaker labels, pauses, overlapping speech notes).

In South African studies, language diversity is critical. Many institutions require ethical compliance when participants speak isiXhosa, isiZulu, Sesotho, Afrikaans, or other languages. A common exam scenario is: the interview is conducted in a local language, but the analysis is done in English. You should discuss translation validity:

• Does translation preserve meaning?
• How are ambiguous phrases handled?
• Who translates (researcher, trained assistant, professional translator)?
• Are back-translation or validation checks used?

2.4 Digital Ethnography and Social Media Data: Opportunities and Pitfalls

Digital ethnography uses online spaces as “field sites.” Social media can provide valuable insights into discourses and community dynamics, especially on topics like activism, service delivery protests, or gender-based violence reporting.

However, exam questions often require you to address:

• Public vs private content: even “public posts” may involve vulnerable users.
• Consent: whether individuals can reasonably be expected to consent to research observation.
• De-identification: anonymising usernames and removing identifiable details.
• Context collapse: online content may be interpreted without the surrounding context.
• Platform policies and data access: limitations on scraping and API usage.

A strong technical-ethical balance answer could say: Use only data collection methods allowed by terms of service and law, store data securely, and focus on aggregated patterns rather than identifiable narratives—unless explicit consent is obtained.

2.5 Geographic Information Systems (GIS) for Social Research

GIS helps connect social outcomes with space, such as:

• access to clinics, schools, or transport routes,
• spatial clustering of service delivery complaints,
• mapping informal settlements and infrastructure constraints (with careful ethical considerations).

Key GIS informatics concepts:

• Geocoding: converting addresses or location descriptions into coordinates.
• Spatial join/overlay: combining layers (e.g., household locations with service coverage).
• Scale and Modifiable Areal Unit Problem (MAUP): results change depending on spatial boundaries used.

Exam-ready reasoning: You should not oversimplify “map = truth.” Spatial patterns can reflect measurement artefacts, boundary choices, or historical data collection biases.

2.6 Case Example: Community Service Access Study in a Johannesburg Area (Illustrative)

Imagine a research project exploring barriers to accessing social services in a Johannesburg locality. The technology plan might involve:

• tablets for interviewer-administered surveys,
• audio-recorded interviews with grant beneficiaries and service providers,
• GPS coordinates for “nearest service point” calculations,
• a secure database for linking survey IDs with interview IDs.

In your exam answer, you would identify:

• how participant codes prevent direct identification,
• how consent is stored separately from responses,
• how GPS data is generalised (e.g., rounding coordinates) to reduce re-identification risk,
• how offline-first tools sync when connectivity returns.

Even though this is illustrative, the underlying logic is what matters: technology choices must be justified by ethics, feasibility, and analytical needs.

3) Data Management, Ethics, Security, and Data Quality for Social Research

A hallmark of Social Sciences Informatics is competence in governance: ethical compliance, secure handling, documentation, and data quality assurance. In assessment situations, marks often reward structured, operational answers: not only “be ethical,” but “how to implement ethical safeguards in the data workflow.”

3.1 Ethical Principles in Informatics-Enhanced Research

Core ethics principles include:

• Respect for persons: informed consent, voluntary participation.
• Beneficence: minimise harm and maximise benefit.
• Justice: fair inclusion and fair distribution of burdens/benefits.
• Confidentiality and privacy: protecting identity and sensitive information.
• Accountability: ability to demonstrate ethical compliance to review boards.

In digital contexts, ethical practice often requires additional steps:

• consent forms that clearly explain audio/video recording,
• privacy explanations about cloud storage or analytics,
• explicit discussion of how long data is stored and who can access it.

3.2 Common Data Protection Risks and How to Mitigate Them

Risks are not abstract; they appear at specific workflow steps.

3.2.1 Re-identification and Data Linking

Even anonymised datasets can become identifiable when combined with other data sources (e.g., rare combinations of demographics, location, and employment status).

Mitigation strategies:

• use participant codes,
• minimise direct identifiers (names, exact addresses),
• generalise or mask precise coordinates,
• remove free-text fields if they include identifying details,
• apply disclosure control for small subgroups.

3.2.2 Loss, Theft, or Accidental Exposure

Devices can be lost; files can be accidentally shared.

Mitigation:

• encryption on devices and storage,
• strong access controls (unique logins),
• auto-lock and secure screen settings,
• secure transfer protocols (encrypted connections),
• regular backups with restricted access.

3.2.3 Insider Threat and Misuse

People with access can misuse data.

Mitigation:

• role-based access control,
• audit logs for data access and file changes,
• separation of duties (e.g., who holds consent records vs who analyses data).

3.3 Data Management Plans (DMP): Turning Ethics into Procedure

A Data Management Plan is a practical governance document. It usually covers:

1. Data types: survey responses, interview transcripts, audio files, GIS layers.
2. Storage locations: secure servers/institutional repositories.
3. Ownership: who owns the data (researchers, institution, participants/community).
4. Retention: how long data is kept.
5. Access conditions: who can access, under what approvals.
6. Backups and recovery: how data is protected against loss.
7. Documentation: codebooks, metadata, version history.
8. Sharing and reuse: whether and how anonymised data can be shared.

A typical exam scenario asks: “What should be included in a DMP for a mixed-methods study?” A well-structured answer lists each data type and how it’s protected.

3.4 Data Security Toolkit: Operational Controls

A “good answer” usually includes both conceptual and operational controls:

• Encryption:
  • at rest (stored files)
  • in transit (data moving between devices)
• Authentication:
  • strong passwords
  • multifactor authentication where available
• Authorisation:
  • least privilege access
• Secure file naming:
  • avoid names; use participant IDs
• Audit trails:
  • track who accessed what and when
• Data minimisation:
  • collect only what is needed
• Secure deletion:
  • when retention ends, follow institutional protocol

3.5 Data Quality: Validity, Reliability, and Missingness

Data quality is a central exam theme. Informatics tools can improve quality, but only if researchers implement quality checks.

3.5.1 Measurement Validity

Validity asks: are we measuring what we think we are measuring?

Examples:

• A “trust” measure might be invalid if questions conflate trust with fear of authorities.
• A “job readiness” scale might be invalid if items do not match local labour market realities.

Mitigation:

• use previously validated scales where appropriate,
• pilot test instruments with local participants,
• revise wording to match local literacy levels.

3.5.2 Reliability

Reliability asks: will the instrument produce consistent results under similar conditions?

• For surveys: internal consistency checks (e.g., Cronbach’s alpha)
• For coding qualitative data: inter-coder reliability or consensus procedures

3.5.3 Missing Data and Bias

Missingness is not neutral. If missing responses correlate with key variables (e.g., younger people answering less), results can be biased.

Key tactics:

• design instruments to reduce avoidable missingness (skip logic, clear options),
• track missingness patterns early,
• decide on missing data handling methods (listwise deletion, imputation) based on assumptions.

3.6 Versioning, Reproducibility, and Documentation

In modern informatics practice, reproducibility matters: other researchers should be able to follow the analysis process and reach consistent outputs.

Practical steps:

• maintain a data dictionary for variable definitions,
• store cleaning scripts and analysis scripts,
• document transformations: recodes, missing value rules, outlier decisions,
• keep dates for each processing step.

An exam-friendly “reproducibility argument” is: if decisions are undocumented, evidence becomes weak because critics cannot evaluate whether bias was introduced through data handling.

3.7 Mini Case: Secure Handling of Audio and Transcripts

Consider a study with 30 interviews. Audio files contain sensitive voice and background context. Transcripts may contain identifying information, such as:

• names of clinics,
• names of persons referenced in stories,
• specific workplaces or addresses.

A robust informatics process:

1. Create participant codes: P001–P030.
2. Store audio files as: audio_P001_2026-03-10.wav.
3. Store consent forms separately as: consent_P001.pdf in a restricted folder.
4. During transcription, redact identifiers or replace with placeholders (e.g., [PERSON_1]).
5. Maintain a “redaction log” describing what changed and why.

The key exam point is that ethics is not only at consent—it continues through digitisation, transcription, and analysis.

4) Analysis with Technology: Quantitative, Qualitative, Mixed Methods, and Responsible Computation

This section builds the bridge from data to findings. It addresses analysis approaches used in technology-assisted social research, including computational supports like coding assistance and mapping, while emphasising that technology must not replace theoretical interpretation.

4.1 Quantitative Analysis: Preparing Datasets for Statistical Reasoning

Quantitative analysis typically begins with a clean, well-documented dataset. Informatics contributions include:

• automated consistency checks,
• variable coding pipelines,
• script-based transformations (so steps are auditable).

Key workflow steps:

1. Codebook creation: variable names, labels, value coding.
2. Data import into statistical software.
3. Cleaning:
   • remove duplicates,
   • handle impossible values,
   • standardise missing values coding (e.g., -99 = “not answered”).
4. Exploration:
   • frequency tables,
   • summary statistics,
   • visualisations.
5. Modeling:
   • regression models,
   • association tests,
   • group comparisons,
   • time series or panel analysis if repeated measures exist.

In exams, you may be asked to justify choosing a model type. For example:

• If your dependent variable is binary (e.g., “received assistance: yes/no”), logistic regression is often appropriate.
• If your dependent variable is continuous (e.g., “perceived service satisfaction score”), linear regression might be considered, subject to assumptions.

4.2 Quantitative Case Scenario: Youth Employment Perceptions

Imagine a CUT-aligned study on youth employment perceptions with a digital survey of 300 participants. Variables include:

• age (years),
• education level (coded categories),
• employment status (unemployed, employed, not in labour force),
• perception score of job opportunity (Likert 1–5),
• confidence in government support (Likert 1–5).

In an exam, a strong answer describes:

• how you’d recode Likert scales (e.g., treat as ordinal or construct composite scores),
• how you’d check distribution and outliers,
• how you’d test whether perceptions differ by education level using ANOVA or appropriate non-parametric tests,
• and how you’d interpret results with limitations.

Technology’s role: statistical tools speed exploration and enable reproducible analysis, but interpretation depends on measurement validity and sampling representativeness.

4.3 Qualitative Analysis: Digital Coding, Transcription, and Trustworthiness

Qualitative analysis often uses technologies for organisation and coding support rather than “automated truth.”

Common steps in technology-assisted qualitative analysis:

1. Transcription and formatting
2. Initial reading and memoing
3. Code development
4. Coding and retrieval
5. Theme building
6. Verification:
   • triangulation,
   • member checking (where feasible),
   • reflexive auditing,
   • peer debriefing.

Software tools may support:

• text search,
• code application,
• linking codes to excerpts,
• visualising code frequency (with caution).

A key exam-ready counter-argument: word frequency is not equivalent to thematic importance. A theme might be conceptually crucial even if not often repeated.

4.4 Mixed Methods: Integrating Quant and Qual with Informatics

Mixed methods combines numerical patterns and interpretive depth. Informatics helps integration through:

• consistent participant identifiers across datasets,
• data alignment tables,
• joint displays (quant and qual summary linked by cases or themes).

Two common integration strategies:

• Convergent design: collect both types around the same time; compare and integrate during interpretation.
• Explanatory sequential design: start with quantitative results and use qualitative interviews to explain “why.”

In exam answers, integration must be explicit. Saying “we did both” is insufficient; you must state how results connect.

4.5 Computational Social Science Supports: From Automation to Responsible Use

Computational methods can include:

• text mining and topic modelling (for large corpora),
• network analysis (friendship networks, communication graphs),
• sentiment analysis (with caution about cultural and language limitations),
• assisted coding (human-in-the-loop).

Responsible use requires acknowledging:

• bias in language models,
• translation errors,
• overinterpretation of outputs,
• lack of transparency in proprietary systems.

In exams, a good approach is to define both opportunities and limits:

• Topic modelling can reveal patterns across large amounts of text.
• But “topics” may be artefacts of preprocessing and algorithm parameters.

4.6 GIS and Spatial Analytics: Statistical Considerations and Ethical Boundaries

When using GIS, quantitative analysis includes:

• spatial autocorrelation (e.g., testing clustering),
• hotspot detection,
• regression with spatial effects (advanced).

Yet, ethical boundaries are crucial:

• mapping vulnerable communities at fine resolution can increase risk of re-identification or harm,
• spatial evidence must not stigmatise communities.

Exam answers should show awareness of how spatial reporting can affect communities.

4.7 Mini Case: Mixed-Methods Study of Education Access Barriers

Suppose the study investigates barriers to education among learners in a specific region.

• Quantitative component:
  • survey on school attendance, transport access, caregiver support, and perceived barriers.
• Qualitative component:
  • interviews with learners, parents, and educators about lived experiences.

In data integration:

• quantitative might show low attendance correlated with transport costs,
• qualitative might explain nuanced barriers: inconsistent transport, unsafe travel routes, and caregiver work schedules.

Informatics helps by enabling:

• linking survey participants to interview themes (via consented identifiers),
• constructing joint displays such as:
  • “Barrier category → quantitative pattern → qualitative explanation.”

5) Designing, Implementing, and Reporting Technology-Enabled Social Research: From Methods to Assessment-Ready Outputs

This final section focuses on exam-critical competencies: producing a coherent research plan, handling technology governance, and writing up outputs that demonstrate methodological excellence. It also includes institution-centered examples relevant to South African education and CUT-aligned expectations about applied research.

5.1 Research Proposal Competence: Aligning Objectives, Methods, Data, and Technology

A strong proposal connects the chain:

Research question → variables/themes → data sources → collection method → technology → analysis → ethical governance

A technology-enabled proposal must also address feasibility:

• Are participants reachable using the proposed technology?
• Does the location have connectivity?
• Do researchers have the devices and training required?
• How will language issues be handled?

A typical exam question may ask: “Critically discuss how technology affects validity and ethics in a study using mobile surveys and audio recordings.”

Your answer should include:

• validity impacts (measurement mode effects, translation issues),
• reliability impacts (standardisation across interviewers),
• ethical impacts (consent for recording, storage security),
• mitigation plans.

5.2 Sampling and Recruitment with Technology: Reducing Bias While Increasing Reach

Technology affects sampling through:

• platform reach (social media may overrepresent some groups),
• recruitment channels (WhatsApp groups may exclude people not in networks),
• self-selection (people who respond may differ systematically).

To strengthen sampling validity:

• consider mixed recruitment channels (community networks + digital),
• monitor demographics of respondents vs target population,
• set inclusion criteria carefully,
• use stratified sampling when possible.

An exam-ready concept is: technology can improve reach, but it can also change the population reached. That must be acknowledged in limitations.

5.3 Instrument Design and Pilot Testing with Digital Tools

Pilot testing is a technical and ethical step. With digital tools, pilot tests help identify:

• broken skip logic,
• unclear wording in mobile layouts,
• slow loading times,
• device compatibility issues,
• audio recording quality problems,
• transcription errors.

A good exam answer should describe pilot outcomes and iterative improvements:

1. Run pilot with a small group representative of target users.
2. Analyse problems:
   • missingness patterns,
   • time per question,
   • user feedback about confusing questions.
3. Revise:
   • wording,
   • response options,
   • instrument structure.
4. Re-check:
   • logic,
   • data capture accuracy,
   • consent clarity.

5.4 Writing Up Results: Evidence Standards in Technology-Assisted Research

Technology changes how results are presented:

• statistical tables with reproducible outputs,
• qualitative themes supported by quotes (with redactions),
• maps and spatial visualisations,
• dashboards for stakeholders.

However, report writing must follow evidence standards:

• explain methods and limitations clearly,
• justify technology choices,
• include data governance measures where relevant,
• avoid overstating findings.

A common exam penalty is overclaiming: “Because we used a machine tool, the results are objective.” A better statement is: the tool supports analysis, but bias is managed through design and governance.

5.5 Reproducible Workflows: Documentation as a Mark-Scoring Asset

Many students underperform by not showing reproducibility thinking. An assessment-ready answer often includes:

• file organisation principles,
• version control concepts,
• documentation of cleaning and coding decisions,
• appendices that include codebooks and transformation rules.

Even if the exam doesn’t require actual code, describing reproducibility practices clearly gains marks.

5.6 Ethical Reporting: What to Mention and What to Omit

Ethical reporting includes:

• consent procedures (especially consent to recording),
• anonymisation approach,
• storage duration and protection,
• whether data will be shared and under what conditions,
• how sensitive data were handled.

Ethical reporting also includes what to omit:

• exact coordinates of vulnerable locations,
• identifiable transcripts,
• direct quotes that include personal addresses or names (unless consent and strong justification exist).

5.7 Cluster/Institution-Oriented Competence: CUT Social Studies Focus Alignment

The course sits inside a “Central University of Technology (CUT) Social Studies Focus” collection. Practically, that means exam performance should show that you can:

• connect informatics methods to social science questions,
• understand research contexts relevant to South Africa,
• write in an applied, evidence-driven way.

In a South African university setting—especially in technology-enabled social research—students are expected to demonstrate competence with:

• ethical review processes and consent documentation logic,
• multilingual and culturally competent instrument design,
• data protection and governance in real conditions (devices, connectivity constraints),
• integration of qualitative and quantitative evidence in a single argument.

5.8 Assessment-Style Templates: How to Structure Answers

You can use the following templates during exams to ensure coherence and mark coverage.

Template A: Technology Choice Justification (4-part)

1. Purpose: What does technology enable for the research question?
2. Method fit: Does it suit the method (survey/interview/GIS/text)?
3. Quality: How does it affect validity/reliability and data quality?
4. Ethics & governance: Consent, privacy, security, storage, anonymisation.

Template B: Ethical Risk Register (short)

• Risk: e.g., re-identification via transcripts
• Likelihood: low/medium/high
• Impact: privacy harm, reputational harm
• Mitigation: redaction, removal of identifiers, secure access
• Residual risk: what remains and how it is communicated

Template C: Mixed Methods Integration Statement (1 paragraph)

• Quantitative shows pattern X.
• Qualitative explains mechanism Y.
• Integration results in a combined interpretation that addresses research question Z.
• Limitations include sampling bias and translation effects.

5.9 Full Example: Mini Research Plan (CUT-Style) for a Social Informatics Assignment

To consolidate learning, here is a coherent mini-plan you could adapt in an exam. It includes the kind of detail expected for SSI21AB.

Topic: Community perceptions of safety and trust in local service actors (e.g., municipal by-law enforcement, community policing forums), using mixed methods and technology-enabled data collection.

1. Research question

   • How do community members explain safety experiences, and what role do trust and service responsiveness play in shaping perceptions?

2. Objectives

   • Measure perceived safety and trust levels across demographic groups.
   • Explore narratives explaining what builds or destroys trust.
   • Identify practical service delivery factors linked to perceptions.

3. Method design

   • Convergent mixed methods:
     • Quantitative survey (tablet-administered)
     • Qualitative interviews (audio recorded)

4. Technology choices

   • Tablets for survey to enforce skip logic and reduce missingness.
   • Audio recorders for interviews for accurate transcription.
   • Simple GIS mapping only at aggregated neighbourhood level for contextual interpretation (no fine-grain identifying maps).

5. Instrument outline

   • Survey sections:
     • demographics,
     • safety experiences frequency,
     • trust scale items,
     • service responsiveness items,
     • open-ended option for additional concerns (with caution for identifiers).
   • Interview guide:
     • experiences of safety incidents,
     • perceptions of enforcement fairness,
     • examples of responsiveness,
     • coping strategies and community dynamics.

6. Sampling and recruitment

   • Recruitment through community associations and enumerator-assisted outreach.
   • Digital access support: enumerators assist participants in survey completion for those with limited device access.
   • Target sample: 300 survey responses and 20 interviews (illustrative plan).

7. Ethics and governance

   • Consent separate from data:
     • Consent forms stored separately in restricted access.
   • Participant coding:
     • audio files named by participant code only.
   • Redaction:
     • transcripts redacted for names and addresses.
   • Secure storage:
     • encrypted device storage and secure server for transfers.
   • Retention:
     • defined retention period per institutional policy.

8. Data processing and analysis

   • Quantitative:
     • clean dataset with codebook and quality checks,
     • compare trust and safety measures across groups,
     • interpret associations cautiously with limitations.
   • Qualitative:
     • thematic coding supported by qualitative software,
     • theme validation through peer checking and memoing.
   • Integration:
     • link quantitative trust patterns to qualitative explanations to answer the research question.

9. Reporting

   • Provide results with methodological transparency.
   • Include limitations:
     • potential mode effects (tablet vs interviewer),
     • translation and transcription impacts,
     • privacy constraints affecting detail in quotes.

This mini-plan demonstrates the course’s core requirement: technology as a research tool under ethical and methodological discipline.

5.10 Common Exam Questions and How to Answer Them

Below are typical question types that students encounter, with guidance on what to include in high-scoring responses.

Q1: “Discuss ethical issues in using digital tools for social research.”

High-scoring answer elements:

• consent specifics (especially for audio/video),
• privacy and anonymisation approaches,
• security controls,
• re-identification risks,
• harm minimisation,
• governance and accountability.

Q2: “Explain how technology can affect validity and reliability.”

Include:

• measurement mode effects,
• instrument usability,
• translation issues,
• coding consistency,
• quality assurance processes.

Q3: “Describe the steps in a data management workflow.”

Include:

• data capture,
• storage and versioning,
• documentation,
• cleaning and processing logs,
• backups and access controls,
• reproducibility documentation.

Q4: “Explain how mixed methods integration works.”

Include:

• design type (convergent/explanatory),
• integration point (during interpretation),
• joint display logic,
• explicit linkage between quantitative patterns and qualitative explanations.

Q5: “Why must researchers be critical of computational tools?”

Include:

• bias, transparency, and interpretability issues,
• cultural/language limitations,
• need for human-in-the-loop validation,
• ethical risks around data use.

Final Consolidation: Core Exam Competencies to Remember

To perform well in CUT SSI21AB: Social Sciences Informatics – Using Technology for Social Research, you should be able to do the following reliably:

1. Define Social Sciences Informatics and distinguish data, information, knowledge, and evidence.
2. Describe a full technology-enabled research lifecycle with risks and mitigations.
3. Design technology strategies that fit methods and contexts (including South African connectivity and language realities).
4. Apply ethical and data governance principles across consent, collection, digitisation, analysis, and reporting.
5. Ensure data quality through validity, reliability, missingness management, and documentation.
6. Perform and justify analyses (quantitative, qualitative, mixed methods), including responsible use of computational supports.
7. Write assessment-ready research outputs that are transparent, reproducible, and ethically reported.

If you can articulate each competency with a structured workflow, operational examples, and clear ethical reasoning, your answers will align with what SSI21AB examinations typically reward: methodological clarity, critical technological literacy, and responsible social research practice.