Evaluation Planning and Commissioning

Introduction

Health-care evaluation is often described as the systematic assessment of the merit, worth, relevance, effectiveness, efficiency, equity, quality, or sustainability of a policy, programme, service, intervention, or organisational change. Yet evaluation does not begin with data collection. It begins earlier, with a disciplined process of deciding whether an evaluation is feasible, why it is needed, who will use it, what questions it must answer, what evidence is credible, and how responsibilities and resources will be governed.

Evaluation planning converts a broad management concern into an assessable and decision-relevant inquiry. Commissioning is the process through which an organisation authorises, funds, specifies, procures, oversees, and uses an evaluation. Poor commissioning can produce technically sophisticated research that does not answer the sponsor’s real questions. Conversely, a well-designed evaluation may be proportionate, practical, ethically sound, and influential even when it uses relatively modest resources.

This chapter examines evaluability assessment, stakeholder analysis, evaluation questions, scope, criteria, governance, resources, timelines, terms of reference, and procurement briefs. It treats evaluation as both an analytical activity and a governance function. In health-care management, this dual perspective is essential because evaluation findings can affect clinical practice, public expenditure, professional reputation, patient safety, access, and trust.

Learning objectives

By the end of this chapter, the learner should be able to:

1. Explain the purpose and components of evaluation planning and commissioning.
2. Conduct a structured evaluability assessment.
3. Identify, analyse, and engage relevant stakeholders ethically.
4. Develop focused evaluation questions, scope, criteria, indicators, and standards.
5. Design proportionate governance arrangements, resource plans, and timelines.
6. Prepare a robust evaluation plan, terms of reference, and procurement brief.
7. Critically assess methodological, ethical, political, and implementation risks.
8. Apply these principles to a realistic health-care commissioning decision.

1. From management concern to evaluation decision

An evaluation should be commissioned because evidence is needed for a decision, not simply because evaluation is fashionable or administratively expected. Common triggers include:

• uncertainty about whether a new service improves outcomes;
• a requirement to decide whether to continue, scale, modify, or stop a programme;
• concern about unequal access or unintended harms;
• accountability for public or organisational resources;
• implementation difficulties;
• conflicting claims from providers, professionals, patients, or policymakers; and
• the need to learn before replicating an intervention elsewhere.

A useful initial distinction is between an evaluation and related activities. Monitoring routinely tracks implementation and performance. Audit assesses compliance against explicit standards. Research seeks to generate generalisable knowledge, although health-care evaluations may also make a contribution to knowledge. Quality improvement uses iterative measurement and change to improve local practice. These activities overlap, but their purposes, governance requirements, and designs are not identical.

The commissioning question should therefore be framed as: “What decision will this evaluation inform, and what evidence is needed before that decision can responsibly be made?” If no realistic decision, user, or action can be identified, commissioning may be premature.

2. Evaluability assessment

2.1 Meaning and purpose

Evaluability assessment is a systematic examination of whether a programme can be evaluated usefully, credibly, and feasibly at a particular point in time. It does not ask whether a programme is successful. It asks whether the programme’s design, implementation, data, context, and decision environment are sufficiently developed to support a worthwhile evaluation.

Evaluability is multidimensional. A programme may be technically evaluable but politically unusable, or theoretically coherent but operationally immature. A defensible assessment considers at least five dimensions:

2.2 Programme theory and logic

A programme theory explains the causal reasoning behind an intervention. A basic logic model connects inputs, activities, outputs, short-term outcomes, longer-term outcomes, and assumptions. For example, a hospital-at-home service may assume that trained staff, remote monitoring, rapid escalation, and coordinated discharge will reduce avoidable inpatient days without increasing adverse events.

The model should distinguish:

• Inputs: funding, staff, equipment, information systems, partnerships.
• Activities: assessment, treatment, education, referral, monitoring.
• Outputs: patients enrolled, visits completed, response times, care plans issued.
• Outcomes: improved symptoms, fewer admissions, better experience, reduced cost.
• Impacts: sustained population health, equity, system resilience, or improved survival.
• Assumptions: conditions believed necessary for the pathway to work.
• External factors: events outside programme control, such as workforce shortages or policy changes.

A logic model is not proof of causation. It is a testable representation of intended relationships. During evaluability assessment, stakeholders should be asked whether the model is clinically plausible, operationally realistic, and sufficiently specific to guide measurement.

2.3 A practical evaluability decision

An evaluability assessment commonly ends in one of four recommendations:

1. Proceed: the programme and evidence systems are sufficiently mature.
2. Proceed with safeguards: evaluation is feasible, but limitations require mitigation.
3. Defer and strengthen: improve implementation, data, or programme definition first.
4. Do not commission a full evaluation: use monitoring, audit, or a focused feasibility study instead.

Deferral is not failure. Evaluating an unstable intervention may produce an apparently negative result that actually reflects incomplete implementation. Equally, indefinite postponement can become a way to avoid accountability. The assessment should therefore specify what must change, who is responsible, and when readiness will be reviewed.

3. Stakeholder analysis and participation

3.1 Stakeholders in health-care evaluation

Stakeholders are individuals, groups, or institutions that affect an evaluation, are affected by the programme, possess relevant knowledge, control resources, experience consequences, or use findings. They may include patients and carers, clinicians, managers, commissioners, regulators, policymakers, professional bodies, community organisations, data controllers, unions, providers, and evaluators.

Stakeholders are not merely audiences for dissemination. They may define the problem, influence access to data, interpret findings, identify unintended effects, and determine whether recommendations are implemented. However, participation must not allow powerful actors to suppress inconvenient evidence or convert evaluation into advocacy.

3.2 Stakeholder mapping

A simple mapping approach assesses stakeholders by influence and interest. This is a starting point rather than a complete analysis.

A stronger analysis also asks: What does each stakeholder value? What evidence do they trust? What risks do they face? What information can they provide? What conflicts of interest exist? Whose perspective is missing? In particular, patient and public involvement should not be limited to those easiest to recruit or most familiar with institutional language.

3.3 Engagement across the evaluation cycle

Engagement can occur during problem definition, theory development, question prioritisation, instrument design, interpretation, dissemination, and implementation of recommendations. Different forms of participation have different purposes. Consultation seeks views; collaboration shares design work; co-production shares decision authority. The chosen level should be explicit.

Participation also has costs. Patients may experience fatigue, distress, or loss of privacy; staff may face workload pressures; community organisations may contribute unpaid expertise. Fair reimbursement, accessible communication, confidentiality, and feedback about how contributions were used are basic requirements of ethical engagement.

4. Defining evaluation purpose, questions, and scope

4.1 Purpose before method

A commissioning brief should state the evaluation purpose in decision terms. For example: “to determine whether the integrated diabetes pathway should be continued and expanded across three additional districts” is more useful than “to evaluate the pathway.” The purpose identifies the decision, population, intervention, setting, and approximate timing.

Evaluation questions should be answerable, prioritised, and linked to evidence. A useful question set may include:

• Descriptive: What was implemented, for whom, where, and under what conditions?
• Process: Was implementation consistent with the intended model?
• Outcome: What changes occurred in health status, experience, access, or service use?
• Causal or contribution: To what extent did the programme contribute to observed changes?
• Economic: What resources were used, and what value was produced?
• Equity: Who benefited, who did not, and whether disparities changed?
• Sustainability: Can benefits and delivery be maintained over time?

Questions should not imply a conclusion. “How effective was the service?” is preferable to “Why was the successful service effective?” unless success has already been established.

4.2 Prioritising questions

An evaluation cannot answer every interesting question. Questions may be prioritised according to decision importance, uncertainty, feasibility, equity relevance, and consequences of error. A sponsor can score each proposed question, for example, from 1 to 5 on these dimensions, but numerical scoring does not replace judgement. The final list should distinguish primary questions from secondary and exploratory questions.

4.3 Scope

Scope defines the boundaries of the evaluation. It should specify:

• intervention or programme components;
• population and inclusion or exclusion criteria;
• geographical and organisational settings;
• time period;
• outcomes and perspectives;
• comparator, if applicable;
• implementation conditions;
• economic and equity dimensions;
• data sources and access arrangements; and
• topics explicitly outside scope.

Uncontrolled scope expansion, or “scope creep,” is a common commissioning failure. A change-control process should require proposed additions to identify implications for cost, time, burden, validity, and governance.

5. Criteria, indicators, standards, and judgement

Evaluation criteria are the dimensions against which a programme is assessed. Common criteria include relevance, effectiveness, efficiency, equity, quality, safety, acceptability, accessibility, appropriateness, coherence, and sustainability. Criteria should be selected because they matter to the decision, not because they create an impressive-looking framework.

An indicator is an observable measure related to a criterion. A standard is the level or condition used to judge performance. For example:

• Criterion: timely access.
• Indicator: median time from referral to first assessment.
• Standard: assessment within 14 days for at least 90% of eligible referrals.

Indicators should have clear numerators, denominators, time frames, populations, data sources, and limitations. A rate can be represented as:

Here, the numerator is the count of events, the denominator is the population at risk, and 10^k is a scaling factor such as 100 or 1,000. The formula is meaningful only when the numerator and denominator are conceptually aligned, the observation period is specified, and the population at risk is correctly defined. A crude rate should not be interpreted as a causal effect or compared across settings without considering differences in case mix and data quality.

Judgement should combine quantitative and qualitative evidence. A target may be met while patients report poor experience, or average improvement may conceal widening inequalities. Evaluation criteria should therefore make trade-offs visible rather than reduce the assessment to a single score.

6. Choosing an evaluation design

Design follows questions, not the preferences of the evaluator or commissioner. Possible approaches include:

• Process evaluation: examines implementation, reach, fidelity, mechanisms, context, and acceptability.
• Outcome evaluation: assesses changes in specified outcomes.
• Impact evaluation: estimates whether observed changes can plausibly be attributed to the intervention.
• Economic evaluation: compares costs and consequences from a defined perspective.
• Realist evaluation: asks what works, for whom, in what circumstances, and why.
• Developmental evaluation: supports adaptation in complex and rapidly changing programmes.
• Mixed-methods evaluation: integrates quantitative and qualitative evidence.

A controlled design may be appropriate where a stable intervention, credible comparator, adequate sample, and ethical equipoise exist. In other settings, contribution analysis, interrupted time series, matched comparison, case study, qualitative comparative analysis, or realist approaches may be more credible and feasible. No design eliminates uncertainty. The plan should explain the assumptions by which evidence will support conclusions.

A mixed-methods design is especially useful where outcomes alone cannot explain implementation. Quantitative evidence may show that emergency admissions changed; interviews, observations, and process data may explain whether the change arose from the intervention, referral patterns, coding changes, or wider system pressures.

7. Governance, independence, and ethics

7.1 Governance architecture

Evaluation governance should define authority, accountability, escalation, data access, quality assurance, and publication rights. A proportionate structure may include:

• Sponsor: owns the decision and provides resources.
• Senior responsible owner: accountable for delivery and organisational support.
• Evaluation steering group: oversees scope, quality, risks, and interpretation.
• Independent evaluator: designs and conducts the evaluation.
• Patient and public contributors: advise on relevance, burden, interpretation, and communication.
• Data custodian or controller: authorises lawful and secure data use.
• Ethics or research governance body: determines required review and safeguards.

The steering group should not rewrite inconvenient findings. Its role is to challenge assumptions, ensure contextual accuracy, and protect integrity. A formal protocol, version control, conflict-of-interest declarations, decision log, and publication policy strengthen transparency.

7.2 Ethics and data protection

Evaluation may involve vulnerable people, sensitive health information, unequal power relationships, and findings that could affect employment or access to services. The plan should address informed consent where required, lawful data processing, data minimisation, secure storage, retention, anonymisation or pseudonymisation, access controls, adverse-event escalation, and accessible communication.

The distinction between service evaluation, research, audit, and quality improvement varies across jurisdictions. Classification should be determined early with the relevant governance authorities rather than assumed by the commissioner. If an activity is likely to produce generalisable research or involves additional procedures, research ethics review may be required.

7.3 Independence and conflicts

Independence does not mean absence of engagement. It means that methods, analysis, conclusions, and reporting are protected from inappropriate influence. Contracts should specify evaluator access to data, rights to report limitations and negative findings, management of sponsor comments, and conditions under which findings can be published. Commercial or institutional interests should be declared and managed.

8. Resources, budgeting, and scheduling

8.1 Resource planning

A realistic budget includes more than evaluator fees. Cost categories may include project management, stakeholder engagement, travel, translation and accessibility, data extraction, software, transcription, statistical or qualitative analysis, patient reimbursement, governance approvals, secure storage, dissemination, and contingency.

Resource estimates should be linked to tasks and deliverables. A useful resource plan identifies the required expertise—for example, clinical, implementation science, health economics, statistics, qualitative research, information governance, and patient involvement—and distinguishes internal contributions from purchased services.

8.2 Timeline logic

Evaluation timelines should reflect programme maturity, recruitment rates, data availability, follow-up periods, approvals, analysis, and decision deadlines. A compressed schedule may be possible for a process review but not for outcomes requiring twelve months of follow-up. Milestones should include initiation, protocol approval, instrument development, pilot testing, data collection, interim review, analysis, draft report, stakeholder validation, final report, and dissemination.

Dependencies should be explicit. For example, analysis cannot begin until data definitions are agreed and data quality is assessed. A decision deadline should not be treated as a substitute for adequate observation time. If an interim report is needed, it should be clear which questions it can and cannot answer.

9. The evaluation plan

An evaluation plan is the operational and methodological blueprint for the work. It commonly contains:

1. Background and rationale.
2. Programme description and theory of change.
3. Purpose and intended users.
4. Primary, secondary, and exploratory questions.
5. Scope and exclusions.
6. Criteria, indicators, standards, and definitions.
7. Design and methods.
8. Sampling and recruitment.
9. Data sources, quality assurance, and analysis.
10. Equity, patient involvement, and accessibility.
11. Ethics, information governance, and safeguarding.
12. Governance, roles, independence, and escalation.
13. Deliverables and reporting arrangements.
14. Resources, budget assumptions, and timeline.
15. Risks, limitations, and mitigation.
16. Dissemination, implementation, and learning arrangements.

The plan should distinguish prespecified methods from adaptations. In complex programmes, adaptation is often necessary, but changes should be documented with their rationale and implications for interpretation.

10. Terms of reference and procurement briefs

10.1 Terms of reference

Terms of reference (ToR) formally describe what an evaluation is commissioned to do. They should be sufficiently specific to permit comparable bids while allowing bidders to propose an appropriate design. Essential components include the commissioning authority, background, purpose, objectives, questions, scope, expected outputs, timetable, budget information, governance, data access, bidder requirements, evaluation criteria for tenders, and submission instructions.

Weak ToR often prescribe a method before clarifying the decision. For example, requiring a survey of exactly 500 patients may be inappropriate if the central problem concerns implementation or if the population is small. Better ToR specify the decision and questions, then invite a justified design.

10.2 Procurement brief

A procurement brief translates the ToR into a fair purchasing process. It should explain:

• the service to be purchased;
• mandatory and desirable requirements;
• expected deliverables and acceptance criteria;
• bidder experience and team composition;
• data security and confidentiality requirements;
• conflict-of-interest declarations;
• pricing format and assumptions;
• evaluation of bids; and
• contract management and variation procedures.

Tender scoring should balance technical quality, methodological fit, relevant experience, equity and engagement capability, delivery feasibility, risk management, and cost. The cheapest bid is not necessarily the best value. A transparent scoring framework reduces arbitrariness and supports auditability.

11. Applied case: evaluating a virtual ward programme

The following is an illustrative case, not a report of an actual organisation.

Background and problem

A regional health system introduced a virtual ward for older adults with chronic respiratory disease. Patients received remote monitoring, telephone review, home visits when necessary, and rapid clinical escalation. After six months, managers observed fewer inpatient bed-days among enrolled patients, but clinicians questioned whether patients had simply been selected because they were at lower risk. Patient groups raised concerns about digital exclusion and caregiver burden. The system had to decide whether to continue and expand the service.

Evaluability assessment

The programme had a documented pathway, trained staff, and electronic records of enrolment and contacts. However, implementation differed across sites, eligibility criteria were inconsistently recorded, and escalation events were not coded uniformly. The assessment concluded that a full impact claim would be premature, but a combined process and outcome evaluation was feasible if measurement was strengthened.

Required preparatory actions included standardising eligibility definitions, recording non-enrolment reasons, defining escalation events, auditing data completeness, and establishing a patient advisory group. A later assessment would determine whether comparative impact analysis was justified.

Stakeholder analysis

High-influence stakeholders included the chief clinical officer, hospital directors, and commissioners. High-interest stakeholders included patients, carers, respiratory nurses, general practitioners, and community providers. Digital-exclusion advocates had less formal power but high relevance to equity. The evaluation therefore included patient interviews, carer input, and analysis by age, deprivation, language, disability, and digital access where lawful and feasible.

Evaluation questions and design

Primary questions were:

1. Was the virtual ward implemented as intended across sites?
2. What changes occurred in admissions, length of stay, patient experience, and safety?
3. For which patients and under what conditions did the model appear beneficial or burdensome?
4. What resources were required, and what are the implications of expansion?

The design combined routine-data analysis, a process evaluation, interviews with patients and staff, observation of referral and escalation processes, and a preliminary cost analysis. The plan avoided claiming that any reduction in admissions was caused solely by the virtual ward because selection bias and concurrent respiratory initiatives remained plausible explanations.

Governance and outcome

An independent evaluation team reported to a steering group that included patient representatives but retained authority over analysis and publication. Interim findings identified inconsistent referral criteria, high value placed on rapid telephone access, and additional workload for some carers. The final recommendation was conditional continuation: expand only after standardising eligibility, improving non-digital routes, monitoring safety, and repeating comparative analysis after twelve months of stable implementation.

Learning points

The case demonstrates that commissioning is a staged decision. Evaluability assessment prevented overclaiming, stakeholder analysis exposed burdens that routine data could not show, and governance arrangements protected credibility. The most useful recommendation was neither immediate expansion nor cancellation, but a specified learning-and-improvement pathway linked to a later decision point.

12. Challenges, limitations, and contemporary developments

Evaluation planning is affected by political urgency, changing interventions, incomplete records, weak data interoperability, staff turnover, and pressure to demonstrate success. Commissioners may unintentionally privilege measurable outcomes over important but difficult-to-measure experiences, relationships, dignity, or equity. Stakeholder consultation can become tokenistic, while co-production can blur accountability if roles are not defined.

Artificial intelligence and automated analytics may assist with record linkage, coding, anomaly detection, forecasting, and qualitative data organisation. They do not remove the need for valid definitions, representative data, human oversight, privacy protection, bias assessment, and transparent documentation. An AI-generated classification may reproduce historical inequalities or obscure why a decision was made. Any use of AI should therefore be specified in the protocol, assessed for data protection and bias, and checked against human-reviewed samples.

Contemporary evaluation increasingly emphasises systems thinking, complexity, equity, implementation, and sustainability. Health interventions interact with professional behaviour, financing, infrastructure, policy, and community conditions. Evaluation plans should therefore identify context and adaptation rather than treating them as noise. At the same time, complexity should not become an excuse for vague questions or untestable claims.

Best-practice checklist

Before approving an evaluation commission, ask:

• Is there a clear decision and identifiable user?
• Is the programme sufficiently defined and mature?
• Is the theory of change plausible and testable?
• Are primary questions prioritised and answerable?
• Are patient, public, workforce, and equity perspectives represented?
• Are criteria, indicators, standards, and data definitions explicit?
• Is the proposed design proportionate to the decision and uncertainty?
• Are independence, ethics, data governance, and publication rights protected?
• Are budget, expertise, access, dependencies, and timelines realistic?
• Do the ToR and procurement brief permit methodological creativity while ensuring comparable bids?
• Is there a plan to use findings, monitor implementation, and revisit decisions?

Chapter Summary

Evaluation planning transforms a broad health-care concern into a credible and usable inquiry. Evaluability assessment examines whether a programme is sufficiently coherent, mature, measurable, feasible, and decision-relevant for evaluation. Stakeholder analysis identifies interests, influence, knowledge, risks, and missing voices, while ethical engagement ensures that participation is meaningful rather than symbolic.

Strong evaluation questions are linked to decisions and may address implementation, outcomes, causation, cost, equity, and sustainability. Scope, criteria, indicators, standards, and data definitions establish what will be judged and how. Design should follow the questions and may combine quantitative, qualitative, economic, realist, developmental, or mixed-methods approaches.

Governance protects accountability, independence, privacy, safety, and integrity. Resource and timeline planning must reflect dependencies and programme maturity. The evaluation plan provides the methodological blueprint; terms of reference define the commission; and the procurement brief supports a fair and auditable purchasing process. Ultimately, a successful evaluation is not simply technically rigorous. It is proportionate, ethical, transparent, interpretable, and connected to a real decision.

Key Terms

Evaluability assessment: A systematic assessment of whether a programme can be usefully and feasibly evaluated.

Programme theory: An explanation of how and why an intervention is expected to produce outcomes.

Logic model: A structured representation linking inputs, activities, outputs, outcomes, impacts, and assumptions.

Stakeholder: A person, group, or institution affected by, influential in, knowledgeable about, or interested in an evaluation.

Evaluation question: A focused question that specifies what the evaluation must determine.

Scope: The defined boundaries of an evaluation, including population, setting, period, outcomes, and exclusions.

Criterion: A dimension used to assess a programme, such as effectiveness, equity, safety, or sustainability.

Indicator: An observable measure related to an evaluation criterion.

Standard: A specified level or condition used to judge performance.

Process evaluation: Assessment of implementation, reach, fidelity, mechanisms, and context.

Outcome evaluation: Assessment of changes in specified outcomes.

Impact evaluation: Assessment of whether changes can plausibly be attributed to an intervention.

Mixed-methods evaluation: Integration of quantitative and qualitative evidence in one evaluation.

Evaluation governance: Structures and processes that allocate authority, accountability, oversight, and safeguards.

Terms of reference: A formal document defining the purpose, scope, requirements, outputs, and arrangements of an evaluation.

Procurement brief: A document that specifies the service to be purchased and the basis for selecting a provider.

Independence: Protection of methods, analysis, conclusions, and reporting from inappropriate influence.

Scope creep: Uncontrolled expansion of an evaluation beyond its agreed boundaries.

Data minimisation: Collecting and retaining only data necessary for a legitimate evaluation purpose.

Contribution analysis: An approach that assesses how evidence supports a reasoned explanation of an intervention’s contribution to observed outcomes.

Review Questions

Short Answer Questions

1. What is the main purpose of an evaluability assessment?
2. Distinguish an evaluation criterion, indicator, and standard.
3. Why should evaluation questions be linked to a decision?
4. Identify four roles that may appear in evaluation governance.
5. What information should be included in a terms of reference document?

Long Answer Questions

1. Explain how a logic model supports evaluation planning, and discuss its limitations.
2. Describe a systematic approach to stakeholder analysis in a health-care evaluation.
3. Compare process, outcome, impact, and realist evaluation approaches.
4. Discuss the ethical and governance issues involved in using patient-level administrative data.
5. Explain how a procurement brief can protect both methodological quality and fairness.

Analytical / Application Questions

1. A new primary-care service has enthusiastic managers but inconsistent implementation and incomplete outcome data. Recommend whether to proceed, defer, or redesign the evaluation, and justify your decision.
2. Design five prioritised evaluation questions for a hospital-at-home programme, ensuring that they address implementation, outcomes, equity, and sustainability.
3. A commissioner receives two bids: one is substantially cheaper but offers limited patient involvement and a rigid survey design; the other is more expensive and proposes mixed methods, stronger governance, and a longer timeline. Develop a decision framework for selecting the better bid.

Multiple Choice Questions with Answers

Question 1

What is the primary purpose of an evaluability assessment?

A. To prove that a programme has achieved its intended outcomes

B. To determine whether a useful and feasible evaluation can be conducted

C. To replace all routine monitoring activities

D. To select the cheapest external evaluator

Correct Answer: B. To determine whether a useful and feasible evaluation can be conducted

Question 2

Which item is an assumption in a logic model?

A. Number of patients enrolled

B. Number of staff trained

C. Belief that trained staff can deliver the intervention effectively

D. Number of consultations completed

Correct Answer: C. Belief that trained staff can deliver the intervention effectively

Question 3

A measure of median referral-to-assessment time is best described as a(n):

A. Criterion

B. Indicator

C. Assumption

D. Governance structure

Correct Answer: B. Indicator

Question 4

Which stakeholder group should generally receive meaningful participation even when it has limited formal influence?

A. Groups affected by the programme and highly interested in its results

B. Only senior executives

C. Only procurement officers

D. Stakeholders with no connection to the programme

Correct Answer: A. Groups affected by the programme and highly interested in its results

Question 5

Which evaluation question is most clearly focused on implementation?

A. Did the programme reduce admissions compared with usual care?

B. Was the programme delivered as intended across participating sites?

C. What is the programme’s cost per quality-adjusted life year?

D. Should the programme be expanded nationally?

Correct Answer: B. Was the programme delivered as intended across participating sites?

Question 6

What is the principal function of a standard in evaluation?

A. To identify the evaluator’s employer

B. To specify the level used to judge performance

C. To guarantee causal attribution

D. To replace the programme theory

Correct Answer: B. To specify the level used to judge performance

Question 7

Which governance arrangement best protects evaluation integrity?

A. Allowing the sponsor to delete unfavourable findings

B. Prespecifying methods and protecting the evaluator’s right to report limitations

C. Preventing patients from reviewing instruments

D. Removing all stakeholder engagement

Correct Answer: B. Prespecifying methods and protecting the evaluator’s right to report limitations

Question 8

According to the chapter’s rate formula, if 24 events occur among 600 people at risk, the rate per 100 people is:

A. 0.4 per 100

B. 4 per 100

C. 24 per 100

D. 40 per 100

Correct Answer: B. 4 per 100

Question 9

Why might a full impact evaluation be deferred?

A. The programme is too successful to study

B. The intervention is immature and implementation or data systems are unstable

C. Stakeholders have been identified

D. A logic model exists

Correct Answer: B. The intervention is immature and implementation or data systems are unstable

Question 10

What is the main purpose of a procurement brief?

A. To report final evaluation findings

B. To specify the service to be purchased and how bids will be assessed

C. To provide patient treatment instructions

D. To replace information-governance approval

Correct Answer: B. To specify the service to be purchased and how bids will be assessed