Unit 3: Analytical and Critical Thinking - Subjective Questions

Analytical reasoning is the systematic process of breaking a complex problem into smaller parts, examining relationships among those parts, and using logic and evidence to reach a conclusion.

It is important in evaluating alternatives because it:

• Clarifies the objectives and constraints of a problem.
• Helps identify the assumptions behind each alternative.
• Enables comparison using criteria such as cost, feasibility, risk, usability, and sustainability.
• Reduces dependence on intuition or personal preference.
• Supports transparent and justifiable decision-making.

In design thinking, analytical reasoning complements empathy and creativity by helping teams select solutions that are both desirable and practical.

Critical thinking involves analyzing information objectively, questioning assumptions, evaluating evidence, and forming reasoned judgments.

When evaluating competing solutions, critical thinking helps a team:

• Define the problem accurately rather than accepting an initial description.
• Distinguish facts from opinions and assumptions.
• Examine the reliability and relevance of evidence.
• Identify strengths, weaknesses, risks, and unintended consequences.
• Consider alternative viewpoints and stakeholder perspectives.
• Reach conclusions that are supported by evidence rather than authority or popularity.

Thus, critical thinking prevents premature commitment to the first attractive idea and encourages a balanced evaluation of all viable alternatives.

Analytical thinking focuses on decomposing information, identifying patterns, understanding relationships, and examining how parts contribute to a whole.

Critical thinking focuses on judging the quality of information, questioning assumptions, testing claims, and deciding whether a conclusion is justified.

The differences can be summarized as follows:

• Analytical thinking asks, "What are the parts, relationships, and possible explanations?"
• Critical thinking asks, "How reliable, logical, and justified are these explanations?"
• Analytical thinking organizes and interprets information.
• Critical thinking evaluates the validity and implications of that information.

They work together when a team first analyzes a problem into its components and then critically evaluates the evidence, assumptions, and conclusions associated with each component.

A systematic evaluation process may include the following steps:

1. Define the decision objective: State what the solution must achieve.
2. Identify evaluation criteria: Include user value, feasibility, cost, time, risk, accessibility, and sustainability.
3. Generate alternatives: Avoid evaluating only the first idea; develop several possible solutions.
4. Collect evidence: Use user research, prototypes, experiments, expert opinions, and available data.
5. Compare alternatives: Use a decision matrix, ranking method, or cost-benefit analysis.
6. Analyze risks and trade-offs: Identify disadvantages, dependencies, and unintended effects.
7. Test promising options: Conduct prototypes or pilot studies with representative users.
8. Make and document the decision: Explain why the selected option best meets the criteria.
9. Review the outcome: Measure performance and revise the solution when new evidence appears.

Cognitive biases are systematic patterns of judgment that can cause people to interpret information or make decisions in ways that depart from objective reasoning.

• Confirmation bias: A person searches for, notices, and remembers evidence that supports an existing belief while ignoring contradictory evidence. For example, a designer who prefers a particular interface may ask users questions that confirm its strengths and overlook negative feedback.
• Anchoring bias: A person relies too heavily on the first information encountered. For example, an initial cost estimate or early design concept may influence later judgments even when better information becomes available.

These biases can be reduced by:

• Defining evaluation criteria before reviewing alternatives.
• Seeking disconfirming evidence.
• Using anonymous idea evaluation.
• Involving diverse team members.
• Repeating tests with independent data.
• Recording assumptions and revisiting them during the project.

• Availability bias: People judge the likelihood of an event based on how easily examples come to mind. A team may overestimate the importance of a rare security incident because it recently received media attention.
• Framing effect: The way information is presented influences a decision, even when the underlying facts are the same. Users may respond differently to a product described as having a "90% success rate" versus a "10% failure rate."
• Sunk-cost fallacy: People continue investing in an unsuccessful project because resources have already been spent. A team may keep improving an unpopular application instead of replacing its basic concept.

These biases can be controlled through neutral wording, statistical evidence, predefined stopping rules, independent reviews, and comparison with realistic alternatives.

A design team can reduce bias by creating a deliberate and transparent decision process:

• Use diverse participants: Different backgrounds and experiences increase the chance that hidden assumptions will be challenged.
• Separate idea generation from evaluation: This prevents early criticism from limiting creativity and prevents attachment to an initial idea.
• Use independent judgments: Team members can score alternatives before group discussion to reduce conformity and groupthink.
• Assign a devil's advocate: One person is asked to present the strongest objections to the preferred option.
• Apply pre-established criteria: Decisions should be based on agreed criteria rather than persuasive personalities.
• Seek disconfirming evidence: Teams should actively look for information that could prove their preferred solution wrong.
• Document assumptions: Written assumptions can be tested and revised.
• Conduct a premortem: The team imagines that the solution failed and identifies possible reasons.

These practices do not eliminate bias completely, but they make reasoning more deliberate and accountable.

Intuitive decision-making relies on experience, pattern recognition, and rapid judgment. It is useful when time is limited and the decision-maker has relevant expertise, but it may be affected by emotion and bias.

Analytical decision-making uses explicit criteria, evidence, comparisons, and logical evaluation. It is useful for high-impact or complex decisions, although it may require significant time and may be limited by incomplete information.

Collaborative decision-making involves stakeholders or team members in defining the problem, generating options, and selecting a solution. It improves diversity of perspectives and stakeholder acceptance, but it can be slow and vulnerable to groupthink.

A strong design process combines these approaches. Intuition can generate hypotheses, analytical methods can test them, and collaboration can improve relevance, acceptance, and implementation.

The rational decision-making model generally follows these steps:

1. Define the problem.
2. Establish objectives and constraints.
3. Generate possible alternatives.
4. Identify evaluation criteria.
5. Gather relevant evidence.
6. Compare alternatives systematically.
7. Select the alternative with the greatest expected value.
8. Implement and monitor the decision.

Its strengths include transparency, consistency, and logical justification. However, complex problems often involve limitations such as:

• Incomplete, uncertain, or contradictory information.
• Difficulty in predicting long-term consequences.
• Conflicting stakeholder values.
• Limited time and resources.
• Criteria that cannot be measured precisely.
• Cognitive and organizational biases.

Therefore, rational analysis should be combined with iterative prototyping, stakeholder feedback, scenario planning, and periodic review rather than treated as a one-time calculation.

Bounded rationality is the idea that people attempt to make rational decisions but are limited by incomplete information, limited time, limited cognitive capacity, and uncertainty about outcomes.

In design thinking, bounded rationality is relevant because:

• Teams rarely know every user need or market condition.
• Future consequences of a solution are difficult to predict.
• It is usually impractical to evaluate every possible alternative.
• Stakeholders may have different interpretations of success.

Instead of seeking a theoretically perfect solution, teams often use satisficing, which means selecting an option that adequately meets the most important requirements. Iterative prototyping, quick experiments, feedback cycles, and staged commitments help teams make reasonable decisions while gradually improving their knowledge.

A decision matrix compares alternatives against a common set of criteria.

The process is:

1. List the alternatives in rows.
2. Define criteria in columns, such as usability, cost, feasibility, risk, and environmental impact.
3. Assign each criterion a weight based on its importance.
4. Score every alternative against each criterion using a consistent scale.
5. Multiply each score by its criterion weight.
6. Add the weighted scores for each alternative.
7. Discuss the results and examine whether the ranking changes when assumptions or weights change.

For alternative , the weighted score can be represented as:

where is the weight of criterion and is the rating of alternative on criterion .

The matrix supports structured comparison, but it does not replace judgment because the choice of criteria, weights, and ratings may contain bias.

A cause-and-effect diagram, also called a fishbone or Ishikawa diagram, helps a team organize possible causes of a problem.

To create one:

• Write the problem or effect at the end of the diagram.
• Create major cause categories, such as people, process, technology, materials, environment, and measurement.
• Brainstorm possible causes under each category.
• Ask repeated "why" questions to move from symptoms to underlying causes.
• Verify important causes using data or observation.

The diagram is useful because it:

• Encourages broad exploration rather than blaming one factor.
• Makes relationships among causes visible.
• Supports team discussion and shared understanding.
• Helps identify areas for further investigation.

It is a hypothesis-generating tool, not proof that every listed cause is valid. Each suspected cause should be tested with evidence.

The 5 Whys technique repeatedly asks why a problem occurred in order to move from an obvious symptom toward a deeper cause. The number five is approximate; questioning should continue until a controllable and evidence-supported cause is identified.

Root-cause analysis is a broader systematic process for identifying the fundamental factors that produce a problem. It may use tools such as the 5 Whys, fishbone diagrams, process maps, fault trees, and data analysis.

They can be used together as follows:

1. Define the problem precisely.
2. Use a fishbone diagram or process map to identify possible causes.
3. Apply the 5 Whys to promising causal chains.
4. Verify the suspected root cause using observations or data.
5. Develop and test corrective actions.
6. Monitor whether the problem is reduced or eliminated.

The combination prevents teams from treating symptoms while also avoiding unsupported assumptions about a single cause.

SWOT analysis examines four categories:

• Strengths: Internal advantages of the proposed solution.
• Weaknesses: Internal limitations or resource requirements.
• Opportunities: External conditions that the solution could use.
• Threats: External risks or conditions that could reduce success.

For example, a digital service may have strong usability as a strength, limited technical capacity as a weakness, growing demand as an opportunity, and new regulations as a threat.

SWOT supports evaluation by:

• Organizing internal and external factors.
• Revealing risks that may not be visible in a purely technical analysis.
• Helping teams identify actions that use strengths to capture opportunities.
• Encouraging preparation for weaknesses and threats.

However, SWOT entries should be supported by evidence and prioritized; an unverified list can become subjective and superficial.

Cost-benefit analysis compares the expected costs of an alternative with its expected benefits. Costs may include development, operation, training, maintenance, and opportunity costs. Benefits may include increased revenue, time savings, risk reduction, improved access, or greater user satisfaction.

A simple net-benefit calculation is:

The method is difficult in social or user-centered problems because:

• Some benefits, such as dignity, trust, inclusion, or well-being, are difficult to express in monetary terms.
• Benefits may occur over a long period and be uncertain.
• Costs and benefits may be distributed unequally across stakeholders.
• Estimates can reflect optimistic assumptions or bias.
• Monetary value may obscure ethical or environmental concerns.

Therefore, cost-benefit analysis should be combined with qualitative evidence, equity analysis, ethical review, and multi-criteria evaluation.

Evidence-based evaluation is the process of judging a solution using relevant, reliable, and systematically collected information rather than unsupported opinions.

Useful evidence may include:

• Qualitative evidence: Interviews, observations, focus groups, diary studies, and open-ended feedback.
• Quantitative evidence: Usage rates, completion time, error frequency, conversion rates, satisfaction scores, and cost data.
• Experimental evidence: Controlled tests, A/B tests, usability experiments, and pilot studies.
• Comparative evidence: Benchmarking against existing solutions or competitors.
• Expert evidence: Assessments from specialists with relevant knowledge.
• Contextual evidence: Environmental, cultural, legal, economic, or accessibility information.

Good evaluation uses evidence that is relevant to the decision, sufficiently reliable, ethically collected, and interpreted in relation to the original objectives.

Validity refers to whether a method actually measures what it is intended to measure. For example, a usability test has high validity when it reflects real user tasks.

Reliability refers to the consistency or repeatability of a measurement. A reliable survey produces similar results under similar conditions.

Relevance refers to how closely the evidence supports the particular decision or problem being considered. Data about a different user group may be reliable but not relevant to the current design.

High-quality evidence should ideally possess all three qualities:

• A measure can be reliable but invalid if it consistently measures the wrong thing.
• Evidence can be valid but unreliable if results vary greatly because of poor measurement conditions.
• Evidence can be valid and reliable but irrelevant to the decision context.

Teams should therefore evaluate the method, sample, conditions, limitations, and connection to the decision before relying on evidence.

Sampling bias occurs when the participants or data selected do not adequately represent the target population. For example, testing an educational application only with highly skilled students may produce overly positive conclusions.

Measurement bias occurs when the method, instrument, wording, or evaluator systematically influences the results. Examples include:

• Leading survey questions.
• A researcher unintentionally influencing interview responses.
• A usability task that does not reflect real use.
• A metric that rewards speed while ignoring accuracy.

These biases can cause teams to select an unsuitable alternative or incorrectly reject a promising one. They can be reduced by:

• Defining the target population clearly.
• Using appropriate and diverse samples.
• Piloting research instruments.
• Standardizing procedures.
• Combining qualitative and quantitative methods.
• Reporting limitations and uncertainty honestly.

Triangulation is the use of multiple sources, methods, investigators, or theoretical perspectives to examine the same question.

For example, a team evaluating a service might combine:

• User interviews to understand experiences.
• Direct observation to identify actual behavior.
• Usage analytics to measure frequency and patterns.
• A prototype experiment to test whether a change improves performance.

Triangulation improves decision-making because it:

• Reduces dependence on the limitations of one method.
• Reveals whether findings are consistent across sources.
• Explains both what happened and why it happened.
• Exposes contradictions that require further investigation.
• Increases confidence in conclusions when independent evidence converges.

Triangulation does not guarantee truth; different methods can share the same bias. The quality and appropriateness of every source must still be assessed.

A prototype is a simplified representation of a product, service, process, or interaction. It allows assumptions to be examined before full implementation.

A structured experiment generally involves:

1. Stating an assumption or hypothesis.
2. Defining the outcome or metric to be observed.
3. Selecting appropriate participants or test conditions.
4. Creating a prototype or test version.
5. Collecting observations and measurements.
6. Comparing results with a baseline or alternative.
7. Interpreting findings and deciding whether to adapt, continue, or abandon the idea.

Prototyping reduces uncertainty by making abstract ideas observable and testable. Low-fidelity prototypes are useful for exploring concepts cheaply, while high-fidelity prototypes are useful for testing detailed interactions. Experiments should be designed to isolate important variables where possible and should acknowledge limitations.