Government Operational Research Service (GORS) Assessment Problem Structuring and Soft OR Methods Syllabus

What does SODA stand for and who developed it?

Strategic Options Development and Analysis, developed by Colin Eden (and Fran Ackermann). It uses cognitive mapping to capture and merge individual and group views to structure complex strategic problems and build commitment to action.

What is a 'cognitive map' in SODA?

A network diagram of concepts (ideas) linked by arrows representing means–ends or cause–effect relationships, often phrased as contrasting poles. It captures how an individual makes sense of an issue; merged maps form a group/strategic map.

In cognitive mapping, how are concepts typically expressed and linked?

Each concept is a short phrase, often with a contrasting pole separated by '...' (e.g. 'expand service ... maintain status quo'). Arrows run from means to ends (cause to effect); a minus sign indicates the link is to the opposite pole.

What is the Strategic Choice Approach (SCA) and who developed it?

A planning-under-uncertainty PSM developed by John Friend and Allen Hickling. It helps groups manage interconnected decisions by working incrementally through commitment and uncertainty rather than seeking a complete plan up front.

What are the four modes of the Strategic Choice Approach?

Shaping (defining the decision problem and links between decision areas), Designing (generating feasible options and combinations), Comparing (evaluating options against comparison areas), and Choosing (managing uncertainty and forming a commitment package). Work moves flexibly between modes.

In the Strategic Choice Approach, what three types of uncertainty are distinguished, and the response to each?

UE — uncertainty about the working Environment (response: more research/surveys); UV — uncertainty about guiding Values (response: clarify objectives/policy guidance); UR — uncertainty about Related decisions (response: coordinate/negotiate with linked decision areas).

In the Strategic Choice Approach, what is a 'commitment package'?

A balanced plan that combines immediate actions and decisions to commit to now, with explorations to reduce uncertainty and contingency arrangements for decisions deferred to later — managing the balance between acting now and keeping options open.

What is a causal-loop diagram (CLD) and what are its building blocks?

A systems-thinking diagram showing how variables influence one another via causal links. Links carry a polarity: positive (+, same direction) or negative (−, opposite direction). Closed chains of links form feedback loops.

Distinguish a reinforcing loop from a balancing loop in a causal-loop diagram.

A reinforcing (R) loop amplifies change — it has an even number (including zero) of negative links and drives growth or decline. A balancing (B) loop counteracts change, seeking stability or a goal, and contains an odd number of negative links.

In a causal-loop diagram, what does a positive (+) link polarity mean versus a negative (−) link?

A '+' link means a change in the cause produces a change in the effect in the SAME direction (both rise or both fall, all else equal). A '−' link means the effect moves in the OPPOSITE direction to the cause.

What is a driver diagram (driver map) and what are its typical three levels?

An improvement/logic tool that links an overall aim to the factors needed to achieve it. Levels: the Aim (goal), Primary drivers (the main areas of influence), and Secondary drivers (specific factors/changes), which then point to interventions/change ideas.

What is an 'issue tree' (or logic tree) and what principle should it follow?

A hierarchical breakdown of a problem or question into progressively finer sub-issues or components. It should be MECE — Mutually Exclusive and Collectively Exhaustive — so branches do not overlap and together cover the whole problem.

What does MECE stand for and why is it useful for structuring problems?

Mutually Exclusive, Collectively Exhaustive. It ensures a breakdown has no overlapping categories (no double counting) and leaves no gaps (covers everything), giving a clean, complete logical structure for issue trees and hypotheses.

What is a logic model and what are its standard components?

A logic model is a linear depiction of how an intervention is expected to work: Inputs → Activities → Outputs → Outcomes → Impact. It links resources to the changes they are intended to produce.

In a logic model, distinguish 'outputs', 'outcomes' and 'impact'.

Outputs are the direct products/deliverables of activities (e.g. people trained); Outcomes are the resulting changes in behaviour/condition (e.g. improved skills, employment); Impact is the longer-term, higher-level change attributable to those outcomes (e.g. reduced poverty).

What is a Theory of Change and how does it differ from a logic model?

A Theory of Change maps the causal pathway from activities to long-term outcomes, making explicit the assumptions and preconditions at each step and often working backwards from the goal. It is richer and more assumption-focused than the typically linear logic model.

What role do 'assumptions' play in a Theory of Change?

Assumptions state the conditions that must hold for one step in the causal chain to lead to the next (e.g. 'trained staff will apply new skills'). Making them explicit allows them to be tested and reveals where the theory might break down.

What is stakeholder mapping/analysis and why is it part of problem structuring?

It is the systematic identification of parties affected by or able to affect an issue, and assessment of their interests, influence and attitudes. It ensures objectives, constraints and acceptability are understood and that engagement is targeted effectively.

Describe the power–interest grid for stakeholder analysis, including the action for each quadrant.

A 2×2 of power (high/low) versus interest (high/low): High power–High interest = Manage closely; High power–Low interest = Keep satisfied; Low power–High interest = Keep informed; Low power–Low interest = Monitor (minimal effort).

In Mendelow's stakeholder matrix, what are the two axes used to classify stakeholders?

Power (the stakeholder's ability to influence the organisation/decision) and Interest (how much they care about or are affected by the issue). The combination dictates the engagement strategy for each stakeholder.

What does the salience model (Mitchell, Agle and Wood) use to classify stakeholders?

Three attributes: Power, Legitimacy and Urgency. Stakeholders are classified by how many they hold — 'definitive' stakeholders possess all three and demand highest priority; those with one attribute (latent) demand least.

What is a prioritisation matrix and what is its general purpose?

A grid that scores or ranks options/issues/drivers against two or more criteria (commonly two dimensions such as impact vs effort, or value vs feasibility) to objectively focus attention on the highest-priority items.

Describe the impact–effort (value–effort) matrix and the label for each quadrant.

A 2×2 of impact (high/low) vs effort (low/high): High impact–Low effort = 'Quick wins'; High impact–High effort = 'Major projects' (big bets); Low impact–Low effort = 'Fill-ins'; Low impact–High effort = 'Thankless tasks' (avoid).

In driver mapping, how do you decide which drivers to prioritise for action?

Prioritise drivers that have high influence/leverage on the aim AND are within your ability to change/control (high controllability). High-influence, high-controllability drivers offer the greatest return on intervention; high-influence but uncontrollable ones need monitoring or risk management.

When scoping an analysis, what should you specify about data and evidence requirements?

What questions/hypotheses the data must answer, what variables/indicators are needed, sources and availability, quality/reliability, granularity and time coverage, gaps and how to fill them, and any access, ethical or quality limitations affecting feasibility.

What factors guide the choice of analytical method for a given problem?

The nature of the question (descriptive, causal, predictive, optimising), the type and availability of data, the degree of agreement on objectives (hard vs soft), time and resource constraints, required rigour, and how results will be used and communicated.

Give examples of when a Soft OR / problem-structuring method is preferable to a quantitative model.

When objectives are unclear or contested, multiple stakeholders disagree, the problem is messy/ill-defined, data is scarce, the issue is political or value-laden, or the main need is shared understanding and buy-in rather than an optimal number.

Why should an analyst make assumptions explicit when presenting a structured approach?

Explicit assumptions make the reasoning transparent and defensible, allow others to challenge or test them, expose where conclusions are sensitive, and prevent hidden errors — all of which strengthen clarity, logic and feasibility of the approach.

When defending a structured approach (e.g. to a client or assessor), what should you be prepared to justify?

Why you framed the question as you did, your choice of scope, objectives and success criteria, why you selected particular methods over alternatives, the assumptions and their risks, feasibility within constraints, and how the analysis answers the real decision.

What is a practical, structured way to tackle an open-ended problem-structuring task under time pressure?

Clarify the real question and client/decision; state objectives, scope, assumptions and success criteria; break the problem into a logical (MECE) structure or hypotheses; identify data/evidence and methods needed; note constraints, risks and stakeholders; then summarise a clear, feasible plan.

What is the value of comparing several Soft OR methods (e.g. SSM vs SODA vs Strategic Choice) when structuring a problem?

Each emphasises something different — SSM clarifies perspectives/worldviews and what the system should do, SODA captures and merges individuals' reasoning into a strategy map, and Strategic Choice manages interconnected decisions under uncertainty — so choosing the right one (or combining them) fits the method to the nature of the messiness.