Identifying Needs & Establishing Requirements (Weeks 10–12) – Comprehensive Study Notes

Week 10: Identifying Needs & Establishing Requirements – Part 1

Aims of the Week

• Enable planning & running of:
  • Successful data-gathering sessions
  • Interviews (face-to-face & remote)
  • Simple questionnaires
  • Observations (direct, indirect, in-the-wild, controlled)

Six Key Issues in Data Gathering

• Setting Goals
  • Decide study objectives and how data will be analysed once collected.
• Identifying Participants
  • Decide who will supply data & how many participants are needed.
• Relationship with Participants
  • Maintain clarity & professionalism.
  • Obtain informed consent when appropriate.
• Ethical Considerations – Collection & Storage
  • Today’s lightweight devices make data capture easy → risk of casual over-collection.
  • Personal data protected by regulations; secure storage mandatory.
• Triangulation
  • Investigate phenomenon from multiple perspectives: different sources, investigators, frameworks, & techniques.
• Pilot Studies
  • Small-scale trial of main study to reveal design or procedural problems.

Capturing Data – Media Choices

• Notes, audio, video, photographs; may be combined:
  • Notes + photos
  • Audio + photos
  • Full video
• Each format brings unique advantages & challenges (storage, richness, analysis time, participant comfort).

Interview Types

• Unstructured: no script → rich, but not replicable.
• Structured: scripted (like questionnaire) → replicable, less rich.
• Semi-Structured: guided by script with freedom to probe → balance of richness & reliability.
• Focus Groups: Group interview; exploit group dynamics.

Interview Question Styles

• Closed questions: predefined answers (e.g., Yes/No) → easier analysis.
• Open questions: free-form answers → depth.

Question-Writing Pitfalls to Avoid

• Long or compound sentences.
• Jargon participants may not know.
• Leading questions (assumptions).
• Unconscious bias (e.g., gender stereotypes).

Interview Structure (Five Phases)

1. Introduction – who you are, study goal, ethics, recording permission, consent form.
2. Warm-Up – easy, non-threatening questions.
3. Main Body – logically ordered core questions.
4. Cooling-Off – few easy questions to defuse tension.
5. Closure – thank, signal end, stop recorder.

Remote Interviews & Focus Groups

• Conduct via Teams/Zoom + collaboration board (Miro).
• Advantages:
  • Participants relaxed in own environment.
  • No travel / dress concerns.
  • Sensitive topics: easier anonymity.
  • Easy voluntary withdrawal.

Enriching Interviews

• Use props (personas, prototypes, scenarios) as prompts.

Questionnaires

• Disseminated online → reach large, unknown populations.
• Closed vs. open questions; closed easier to analyse & computer-score.
• Sampling problem: population size often unknown online.

Questionnaire Design Guidelines

• Question order impacts response.
• Different versions may be needed for different groups.
• Clear completion instructions.
• Keep length reasonable; allow staged opt-outs.
• Pay attention to layout & pacing.

Question & Response Formats

• Closed responses:
  • Radio buttons (single).
  • Check boxes (multiple).
  • Rating scales: Likert, semantic differential, $3, 5, 7$ or more points.
• Open-ended responses.

Encouraging High Response Rates

• Clarify purpose; promise anonymity.
• Pilot test.
• Offer short version.
• Follow-up reminders.
• Provide incentives (e.g., vouchers).
• $40\%$ response rate typically acceptable; lower common.

Administering Online Questionnaires

1. Plan timeline. 2. Design offline. 3. Program/enter template. 4. Test behaviour. 5. Pilot with non-sample users. 6. Recruit participants.

Observation Techniques

• Direct – In the Wild:
  • Use structuring frameworks; decide participation degree (passive ↔ participant); ethnography possible.
• Direct – Controlled:
  • Think-aloud (participants verbalise thoughts while acting).
• Indirect:
  • Diaries, interaction logging, web analytics, data scraping, remote videos/photos, wearables, social media.

Structuring Frameworks (Robson & McCarten, $2016$)

• Simple: Person–Place–Thing (Who? Where? What?).
• Detailed $\rightarrow$ Space, Actors, Activities, Objects, Acts, Events, Time, Goals, Feelings.

Planning Observations in the Wild

• Decide role (passive–active).
• Gain acceptance, respect culture & private spaces.
• Plan: what data & equipment, when to stop.

Ethnography

• Philosophy + techniques (participant observation, interviews).
• Immersion in participants’ culture; participation degree varies.
• Continuous data analysis; questions refined iteratively.
• Requires cooperation; reports rich with examples.

Materials Collected in Ethnography (Crabtree $2003$)

• Activity descriptions, rules, talk recordings, informal interviews, layout diagrams, photos/videos of artefacts, workflows, process maps.

Think-Aloud Technique Example

• Participant verbalises: typing URL, searching, interpreting results, internal deliberations.

Putting Techniques to Work

• Choice influenced by study focus, participants, technique nature, resources.
• Techniques often combined; adapt for different participant needs (e.g., Likert faces for children, GPS tracker on a cat).

Best Practices for Remote Data Gathering (Mastrianni $2021$)

• Establish remote access; include in IRB; pilot test; have backups.
• Inform participants of tech requirements; use familiar tools; consider retrospective questioning if think-aloud fails.
• Define researcher roles; introduce team at session start.

Key Points Recap (Week 10)

• Sessions need clear goals; may require consent.
• Six key issues frame planning.
• Data capture media combinations.
• Interview, questionnaire, observation distinctions.
• Techniques often combined & adapted.

Week 11: Identifying Needs & Establishing Requirements – Part 2

Overview & Purpose

• What/How/Why of requirements.
• Data gathering for requirements.
• Bringing requirements to life via personas & scenarios.
• Capturing interaction with use cases.

Purpose of Requirements Activity

• Explore problem space & define design challenge.
• Represent requirements in prototypes, rigorous notations, acceptance criteria, etc.

Importance – “Miscommunication Cartoon”

• Stakeholders interpret the same idea differently (customer, project leader, analyst, programmer, installer, etc.). Requirements phase is where miscommunication most often occurs.

Definition of a Requirement

• Statement describing what a product should do or how it will perform.

User Stories (Agile)

• Format: As a , I want so that .
• Example: As a traveler, I want to save my favorite airline so that I collect air miles.

Volere Shell Example (Req #$75$)

• Elements: Requirement type, description, rationale, source, fit criterion, satisfaction/dissatisfaction scales, dependencies, conflicts, history.

Categories of Requirements

1. Functional – system behaviours.
2. Data – storage needs & characteristics.
3. Environment / Context of Use
   • Physical (dust, noise, heat).
   • Social (collaboration, privacy).
   • Support (training, comms).
   • Technical (platform compatibility).
4. User Characteristics
   • Background, abilities, usage levels (novice, expert, casual, frequent).
   • Design implications: novices need prompts; experts need power; frequent users want shortcuts.
5. Usability Goals & User Experience Goals.

Usable Security Example

• Robust security without harming UX.
• Visibility of mechanisms, password strength sonification, design trade-offs.

Seven Product Dimensions (Gottesdiener & Gorman, $2012$)

• User, Interface, Action, Data, Control, Environment, Quality Attribute.

Data Gathering for Requirements

• Interviews, observation, questionnaires.
• Studying documentation (procedures, rules, legislation) – useful but not sole source; saves stakeholder time.
• Researching similar products – inspiration & requirement prompts.

Combining Techniques – Case Examples

• Multiple devices: direct + indirect observation, interviews, diaries, surveys (Hollis $2017$).
• Traumatic brain injury aid: interviews, think-aloud, questionnaire, prototype eval.
• Ship manoeuvring system: docs, system eval, user observation, focus groups.
• Smart meters: questionnaire, focus group, design probe, user study.

Probes with Stakeholders

• Design probe – tailored artefact for context.
• Cultural probe – postcards, maps, cameras, diaries.
• Technology probe – working prototype in real context.
• Provocative probe – challenge norms.

Contextual Inquiry (CI)

• One-on-one field interview ($1.5{-}2$ hours) with master–apprentice stance.
• Principles: Context, Partnership, Interpretation, Focus.
• Uses “Joy of Life” & “Joy of Use” concept lists; interview in four parts (overview, transition, main, wrap-up); followed by interpretation session creating contextual design models.

Brainstorming for Innovation (Osborn’s 4 Rules, $1930s$)

• Quantity over quality; defer criticism; encourage wild ideas; combine & improve ideas. Requires facilitation.

Bringing Requirements to Life

• Personas – archetypal users synthesized from data; not real individuals.
• Scenarios – informal narratives describing persona interacting with system.
• Relationship: Persona (who) ↔ Scenario (story) ↔ Goal (motivation).

Persona Examples

• Lena (50, civil servant, dual smartphones, Apple laptop) – commuting, charging issues, techno usage profile.
• Will (35, plumber, family traveller) – needs comprehensive, family-friendly booking, hates disparate systems.

Scenario Example – Group Travel Organizer

• Thomson family collaboratively exploring Mediterranean sailing, using system from multiple locations/devices, negotiating children’s concerns, saving options for next day.

Scenarios Media Forms

• Text, audio, video, animations (e.g., Nilsson $2020$ UbiComp visions).

Design Fiction vs. Scenario

• Design fiction: explores future tech ethically/emotionally; quest-like narrative (e.g., privacy, surveillance).
• Scenarios focus on overcoming a specific “monster” (problem).

Use Cases

• Capture functional interactions.
• Essential use cases: abstract division of user & system intentions.
• Detailed use cases: normal + alternative flows.

Essential Use Case Example – retrieveVisa

• User Intention: find requirements, choose format, obtain copy.
• System Responsibility: request info, supply data, offer formats.

Detailed Use Case & Alternatives – Travel Organizer

• Steps 1–9 normal flow (ask country, validate, ask nationality, provide visa info, offer sharing).
• Alternative flows handle invalid country, invalid nationality, missing visa data (error messages, loop back).

Summary (Week 11)

• Clear requirement statements avoid miscommunication.
• Categories: functional, data, environmental, user, usability, UX.
• Personas & scenarios humanize needs, used throughout lifecycle.
• Use cases detail interactions.

Week 12: Identifying Needs & Establishing Requirements – Part 3

Goals of the Week

• Distinguish qualitative vs. quantitative data & analyses.
• Analyse data from questionnaires, interviews, observations.
• Introduce supporting software (spreadsheets, R, SPSS, Nvivo, Dedoose).
• Identify pitfalls; present findings meaningfully.

Quantitative vs. Qualitative Data & Analysis

• Quantitative data: numbers.
  • Analysis: numerical methods, sizes, magnitudes.
• Qualitative data: words, images; may convert to numbers but not always meaningful.
• Caution: manipulation of numbers can mislead.

Basic Quantitative Techniques

• Averages:
  • Mean: $\bar{x}=\dfrac{\sum x_i}{n}$
  • Median: middle ranked value.
  • Mode: most frequent value.
• Percentages; graphs to show patterns/outliers.
• Question design influences analysis (open vs. closed; fixed alternatives restrict findings).

Basic Qualitative Techniques

• Coding – central.
  • Inductive (bottom-up) vs. Deductive (top-down).
  • Codes must be meaningful & non-overlapping; choose proper granularity.
• Identifying Themes – emergent; often inductive.
• Categorizing Data – deductive scheme; may mix with inductive.

Analytical Frameworks

• Conversation Analysis – micro-level management of talk.
• Discourse Analysis – how language constructs meaning; uncovers hidden assumptions.
• Content Analysis – classify into themes & count frequencies across any media type.
• Interaction Analysis – understand interactions between people & artefacts using video, team interpretation.
• Grounded Theory – build theory through open, axial, selective coding; iterative comparisons.
  • Example: incremental game analysis (Alharti $2018$).
• System-Based Frameworks – socio-technical systems theory, distributed cognition; handle large heterogeneous data (e.g., hospital, airport).

Choosing a Framework (Comparative Table Highlights)

• Input data types, focus, expected outcomes, granularity range from word-level to organizational macro-level.

Tools Supporting Analysis

• Spreadsheets – quick stats & graphs.
• Statistical packages – R, SPSS for deep quantitative work.
• Qualitative software – Nvivo, Dedoose; CAQDAS network resources.

Interpreting & Presenting Findings

• Use visualizations (e.g., pie charts of mobile app usage, session timelines).
• Structured notations (e.g., use cases) convey precise viewpoints.
• Stories/narratives communicate insights intuitively.
• Summaries may combine multiple notations for clarity.

Common Pitfalls

• Drawing causal conclusions from descriptive stats.
• Over-aggregating qualitative nuances.
• Losing context when using software tools.
• Misleading graphs (scale tricks).
• Treating percentages without considering sample size.

Key Points Recap (Week 12)

• Analysis depends on original data-gathering technique.
• Qual & quant data can stem from any approach.
• Means, medians, modes can diverge → choose appropriately.
• Graphs reveal patterns & outliers quickly.
• Qual analysis mixes inductive & deductive coding.
• Multiple frameworks exist, each suited to different data granularity & goals.