Timedora

3 Criteria and Methods for Comparing Options

Learn how to compare policy options using relevant criteria, transparent priorities, suitable appraisal methods, and reviewable evidence.

Purpose and baseline

Comparing policy options is a structured way to assess how well each alternative meets a defined need and what trade-offs achieving it may require. The comparison should inform a decision rather than present value judgments as objective facts. Include a baseline, such as continuing current policy, so proposed options can be judged against what would otherwise happen.

Select and define criteria

Begin with policy objectives and the concerns of affected groups. Turn each objective into a : a specific aspect of performance that can be used to compare alternatives. Criteria may address:

  • Effectiveness: how much an option achieves the intended outcome.

  • Equity: who benefits, who bears costs, and whether impacts are distributed fairly.

  • Cost and value: what resources are required and what outcomes are achieved for them.

  • Feasibility: whether implementation is possible with available authority, staffing, time, and capacity.

  • Risk and wider effects: uncertainties, unintended consequences, and environmental or social impacts that may arise.

Define each precisely and specify the evidence or indicator to use. For example, affordability might mean annual public expenditure, while equity might mean the change in service access for a specified underserved group. Select criteria that matter to the decision and avoid counting the same effect more than once. The criteria selected shape which impacts are visible in the comparison, a point emphasized in OECD guidance.

Set priorities and thresholds

Separate from criteria that involve trade-offs. A legal obligation or essential safety standard can be a pass/fail threshold: an option that fails it should not be rescued by a high score elsewhere. Other criteria can be weighed against one another once options meet the basic requirements.

Criteria are rarely equally important. Explain how priorities were set, who contributed, and where views differ. With , compare the practical change from the worst to the best performance on each , then judge how valuable that change is. A can be important in general yet do little to distinguish options if they perform similarly on it. UK Treasury Green Book guidance supports MCDA, including , for complex trade-offs in longlist appraisal.

Weights express judgments; they are not measurements of universal importance. Document their rationale and, where relevant, show different stakeholder perspectives rather than forcing artificial agreement. A weighted total should neither override a mandatory threshold nor conceal a serious impact on a particular group.

Choose a comparison method

Choose a method that fits both the decision and the available evidence:

  • Threshold screening removes options that fail defined legal, safety, or minimum-performance requirements.

  • estimates and compares costs and benefits in monetary terms where this can be done credibly. It can help show whether an option’s benefits justify its costs, but some effects may be difficult to value in money.

  • compares the cost of options pursuing a shared outcome, such as cost per additional person served. It is especially useful when the outcome can be measured consistently but not reliably monetized.

  • organizes options against multiple criteria, including qualitative and quantitative ones, and makes trade-offs explicit. A performance matrix can support discussion, but a ranking or combined score should not be treated as the decision itself.

A simple weighted-score table can look clear while hiding arbitrary scales or weights. The Green Book distinguishes MCDA from simple “weighting and scoring” and cautions against relying on the latter because it may lack an objective basis and reduce transparency. At the shortlist stage, use the appropriate detailed appraisal, such as CBA, rather than treating MCDA as its substitute.

Compare options in practice

Consider a health agency comparing text reminders, transport vouchers, and mobile clinics to reduce missed appointments. It might define indicators for attendance, access among underserved residents, public cost, and staffing or delivery demands. The comparison could describe the evidence for each option against each indicator, note uncertainty, and identify any .

Text reminders may be inexpensive and easy to deploy but do little for people without reliable phone access. Vouchers may improve access for some patients but require eligibility and payment systems. Mobile clinics may reach people facing the greatest barriers, while costing more and requiring additional staff. These trade-offs do not automatically imply choosing the cheapest option or the one with the top score. Decision-makers should consider the size and distribution of expected effects, evidence quality, and the trade-offs they are willing to accept.

Make the conclusion reviewable

Make the comparison reviewable by stating the options and baseline, defining criteria and indicators, identifying data sources and assumptions, and explaining how priorities and trade-offs were handled. Show uncertainties and impacts across relevant groups. OECD guidance emphasizes transparent comparison and public scrutiny of the criteria, data, and reasoning used to select a preferred option.

Use to test whether the conclusion changes when key assumptions, weights, or forecasts change. If a small change reverses the ranking, report that the choice is sensitive rather than presenting one ranking as certain. A defensible comparison makes clear not only which option appears preferable, but why and under what conditions.