Foundations and Scope of Health Care Evaluation

Introduction: Why Evaluation Matters

Health care organizations make decisions under conditions of uncertainty. They introduce clinical programs, redesign services, purchase technologies, train professionals, regulate providers, and allocate scarce resources. Each decision creates a need for credible evidence: Was the intervention implemented as intended? Did it improve care? For whom did it work, under what conditions, and at what cost? Were unintended harms produced? Evaluation provides systematic ways to answer these questions and to support learning, accountability, improvement, and public stewardship.

Health care evaluation is broader than measuring patient outcomes. It examines the relationship between needs, resources, activities, implementation, outputs, outcomes, impacts, costs, equity, safety, and context. A technically strong evaluation may still be of little practical value if it addresses the wrong question, ignores stakeholders, arrives too late for decisions, or reports results in a form that cannot be used. Conversely, a modest evaluation can be highly valuable when it is purposeful, proportionate, ethically conducted, and closely connected to management and clinical action.

Evaluation is therefore both an analytical activity and a governance practice. It links evidence to judgment and judgment to action. This chapter establishes the conceptual foundation required for later study of evaluation designs, indicators, data collection, economic evaluation, quality measurement, and implementation assessment.

Learning Objectives

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

1. Define health care evaluation and distinguish it from related forms of inquiry.
2. Explain the purposes, stakeholders, and contexts of evaluation.
3. Distinguish monitoring, audit, assessment, research, quality improvement, and program evaluation.
4. Explain formative, process, outcome, impact, and summative evaluation.
5. Select an evaluation approach that fits a decision, intervention, setting, and stage of development.
6. Identify ethical, equity, governance, and practical issues that influence evaluation quality.

The Meaning of Health Care Evaluation

Evaluation is the systematic collection, analysis, and interpretation of information to judge the merit, value, performance, implementation, or effects of an object of evaluation. The object may be a policy, program, service, clinical pathway, organization, technology, workforce initiative, public health intervention, or financing arrangement.

Three features are central.

First, evaluation is purposeful. It is undertaken to inform a decision or judgment. The purpose may be improvement, continuation, expansion, redesign, accountability, accreditation, resource allocation, or learning.

Second, evaluation is systematic. It uses an explicit question, a defensible design, appropriate data, transparent reasoning, and criteria for interpretation. Systematic does not necessarily mean highly quantitative. Qualitative interviews, observations, document analysis, and participatory methods can be systematic when they are planned and rigorously conducted.

Third, evaluation involves judgment. Data alone do not determine value. A waiting time of 45 minutes may be acceptable in an emergency department during a major incident but unacceptable for a scheduled outpatient service. Interpretation requires standards, comparisons, stakeholder values, context, and attention to trade-offs.

A useful general expression is:

Evidence + criteria + context → evaluative judgment → decision or action

This is not a statistical formula. It is a conceptual model. Evidence describes what happened; criteria indicate what is considered desirable; context explains why results occurred and whether they are transferable; judgment integrates these elements for action.

Evaluation questions

A well-designed evaluation begins with a decision-oriented question. Examples include:

• Is a telehealth service reaching the intended population?
• Is a new medication-reconciliation process being implemented reliably?
• Has a maternal-health program reduced avoidable complications?
• Should a pilot service be expanded across a health system?
• Does a screening program produce sufficient benefit relative to its harms and costs?
• How does an intervention affect disadvantaged groups compared with the overall population?

Questions should specify the object, population, setting, time frame, outcomes of interest, and intended use of findings. Vague questions such as “Is the program successful?” usually produce unfocused data and weak conclusions.

A Logic Model for Evaluation

A logic model represents how an intervention is expected to produce results. A conventional sequence is:

Needs and context → Inputs → Activities → Outputs → Short-term outcomes → Longer-term outcomes and impacts

• Needs and context include disease burden, service gaps, population characteristics, policy conditions, organizational culture, and competing interventions.
• Inputs include funding, staff, facilities, equipment, information systems, partnerships, and leadership.
• Activities are the actions undertaken, such as consultations, vaccination, training, referral, or case management.
• Outputs are immediate products of activities, such as number of consultations, staff trained, or patients contacted.
• Outcomes are changes in knowledge, behavior, access, clinical status, experience, or service performance.
• Impacts are broader and generally longer-term effects, such as reduced mortality, improved population health, reduced inequity, or increased system resilience.

The model should not be treated as a guaranteed causal chain. External events, implementation variation, patient characteristics, and other policies may influence every stage. A logic model is best used as an explicit statement of assumptions that can be examined through evaluation.

Figure: A logic model for linking health care needs, implementation, results, and impact.

Purposes of Health Care Evaluation

Evaluation commonly serves several overlapping purposes.

Improvement and learning

Evaluation helps practitioners identify bottlenecks, understand variation, test adaptations, and refine services. Learning-oriented evaluation is particularly important when interventions are complex or implemented in changing environments.

Accountability and stewardship

Public agencies, boards, funders, regulators, and citizens need evidence that resources are used responsibly and that services meet obligations. Accountability evaluation may examine legality, efficiency, safety, access, financial performance, and achievement of stated commitments.

Decision support

Managers may need to decide whether to continue, modify, scale, integrate, or discontinue an intervention. Evaluation makes assumptions and trade-offs more visible.

Knowledge development

Some evaluations contribute to broader knowledge about how interventions work, for whom, and under what conditions. However, an evaluation designed for local management may not support generalization to other settings without additional methodological work.

Equity and voice

Evaluation can reveal whether benefits and burdens are distributed fairly. It can also include the perspectives of patients, carers, communities, and staff whose experiences may be missed by routine administrative data.

These purposes can conflict. A funder may want a definitive judgment, while frontline staff may need rapid feedback for adaptation. An evaluation should make these purposes explicit rather than pretending that one design can meet every need equally well.

Stakeholders and Intended Users

Stakeholders are individuals, groups, or institutions affected by an intervention or interested in its evaluation. Typical stakeholders include:

• patients, service users, carers, and communities;
• clinicians, managers, allied health professionals, and support staff;
• providers, professional bodies, and labor representatives;
• health departments, regulators, commissioners, and accreditation agencies;
• funders, insurers, donors, and taxpayers;
• technology suppliers and implementation partners;
• researchers and academic institutions; and
• populations that may be excluded or disadvantaged by the service.

Stakeholder participation is not merely a courtesy. Stakeholders help define relevant questions, identify feasible indicators, interpret unexpected findings, and judge whether recommendations are usable. Nevertheless, participation must be managed carefully. Powerful stakeholders should not dominate the evaluation, and involvement should not compromise independence or confidentiality.

An evaluation plan should identify primary users and specify the decisions they may make. It should also distinguish users from data providers and subjects. A patient may provide data but have no authority over the final decision; conversely, a regulator may use findings without having participated in data collection.

Contexts of Evaluation

The same intervention can produce different results in different contexts. Important contextual dimensions include:

• Organizational context: leadership, staffing, culture, communication, governance, and competing priorities.
• Clinical context: disease severity, referral patterns, professional norms, and availability of complementary services.
• Social context: language, trust, poverty, transport, health literacy, discrimination, and community support.
• Policy and regulatory context: reimbursement, privacy law, licensing, national standards, and political priorities.
• Technological context: interoperability, data quality, digital access, algorithmic performance, and cybersecurity.
• Temporal context: implementation stage, epidemics, labor disruption, economic change, or policy transition.

Context is not simply background information. It may modify the effect of an intervention or explain why implementation succeeded in one site and failed in another. Evaluation should therefore record contextual conditions and avoid interpreting results as if the intervention operated in isolation.

Distinguishing Evaluation from Related Activities

The terms used in health care organizations often overlap. Clear distinctions improve design and communication.

Monitoring

Monitoring is the routine and continuing observation of activities, outputs, and selected outcomes. Examples include tracking bed occupancy, vaccination coverage, stock-outs, readmissions, or complaints each month. Monitoring is often descriptive and does not, by itself, establish why a change occurred or whether an intervention caused it. Monitoring data may trigger an evaluation when an unexpected trend appears.

Audit

Audit compares actual practice, records, or performance with explicit standards. A clinical audit may examine whether eligible patients received recommended prophylaxis. Audit is usually concerned with conformity and corrective action. It may form part of an evaluation, but an audit alone does not necessarily assess broader effectiveness, unintended effects, or causal contribution.

Assessment

Assessment describes a condition or determines needs, risks, readiness, capacity, or feasibility. A community health-needs assessment may identify disease burden and barriers to care before a program is designed. Assessment often supplies baseline information for later evaluation.

Research

Research aims to generate or contribute to systematic knowledge. It may be exploratory, explanatory, descriptive, or evaluative. The distinction between research and evaluation is not absolute. A local evaluation can produce generalizable knowledge, and research can inform local decisions. The practical distinction usually concerns primary purpose, intended users, governance requirements, and the extent of generalization.

Quality improvement

Quality improvement uses structured change methods to improve care in a particular setting. Teams may use iterative cycles, process mapping, run charts, root-cause analysis, or testing of small changes. Quality improvement is action-oriented and often accepts adaptation during implementation. Evaluation may be embedded within it to determine whether improvement occurred and whether it can be sustained.

Program evaluation

Program evaluation systematically examines a program’s rationale, implementation, outputs, outcomes, impacts, efficiency, equity, and sustainability. It usually asks a broader judgment question than monitoring or audit and a more context-specific question than conventional research. Strong program evaluation may combine quantitative and qualitative evidence.

Types of Evaluation by Timing and Function

Evaluation categories are not mutually exclusive. A single project may include several types.

Formative evaluation

Formative evaluation occurs before or during early development. It examines needs, acceptability, feasibility, design quality, cultural fit, accessibility, and likely barriers. Methods may include literature review, stakeholder consultation, usability testing, pilot studies, focus groups, and prototype testing.

For example, before launching an online mental-health service, formative evaluation might test whether users understand the registration process, whether the language is acceptable, and whether people with disabilities can navigate the platform.

Process or implementation evaluation

Process evaluation examines what was delivered, to whom, with what fidelity, intensity, reach, quality, and adaptation. It asks whether an apparently ineffective program failed because its theory was wrong or because it was not implemented adequately.

Important process dimensions include:

• Reach: whether intended groups received the intervention.
• Dose or intensity: how much intervention participants received.
• Fidelity: the extent to which core components were delivered as designed.
• Adaptation: justified changes made to fit local circumstances.
• Acceptability: how stakeholders perceive the intervention.
• Feasibility: whether the intervention can be delivered with available resources.
• Sustainability: whether benefits and activities can continue over time.

Fidelity should not be treated as rigid uniformity. Some adaptations preserve the intervention’s active ingredients while improving contextual fit.

Outcome evaluation

Outcome evaluation examines short- and medium-term changes plausibly associated with an intervention. Outcomes may include symptom control, functional status, patient experience, access, safety, adherence, service utilization, or staff capability.

Outcome evaluation requires careful definition of the population, comparison, time horizon, and measurement method. A change in an outcome is not automatically evidence of program effectiveness because secular trends, case-mix changes, and other interventions may also explain it.

Impact evaluation

Impact evaluation focuses on longer-term or broader effects and places stronger emphasis on causal attribution or contribution. Examples include population mortality, health inequalities, long-term disability, economic consequences, and system resilience.

In complex health systems, attribution may be difficult. Multiple programs interact, outcomes develop slowly, and controlled comparisons may be infeasible or unethical. Contribution analysis, realist evaluation, interrupted time-series designs, natural experiments, and mixed-method approaches can help explain how an intervention plausibly contributed to observed change.

Summative evaluation

Summative evaluation provides an overall judgment at a defined point, often concerning effectiveness, value, efficiency, equity, or readiness for continuation or scale-up. It may draw on formative, process, outcome, and impact evidence.

A summative judgment should state the criteria used. “Successful” could mean clinically effective, cost-effective, equitable, acceptable, sustainable, or compliant; these are not interchangeable.

Evaluation Frameworks and Theories

The theory of change

A theory of change explains why and how activities are expected to produce results. It identifies causal assumptions, intermediate steps, risks, and external conditions. It is especially useful when interventions involve behavior change, organizational transformation, or multiple levels of action.

A theory of change should be testable. Statements such as “training will improve care” should be unpacked: training must be attended, understood, retained, supported by supervision, translated into behavior, and compatible with staffing and workflow. Evaluation can examine each link.

The CIPP framework

The CIPP framework organizes evaluation around Context, Input, Process, and Product. Context evaluation examines needs and problems; input evaluation examines strategies and resources; process evaluation examines implementation; product evaluation examines results. Its value lies in supporting decisions throughout a program’s life rather than waiting for a final verdict.

Realist evaluation

Realist evaluation asks, “What works, for whom, in what circumstances, and how?” It emphasizes relationships between context, mechanisms, and outcomes. A reminder system may work when staff trust the data and have time to respond, but fail where alerts are excessive or responsibilities are unclear. Realist reasoning is useful for complex interventions that cannot be understood through average effects alone.

Utilization-focused evaluation

Utilization-focused evaluation begins with intended users and intended uses. Design, data collection, analysis, and reporting are shaped by the decisions that findings must support. This approach does not reduce rigor; it directs rigor toward useful questions and feasible decisions.

Donabedian’s structure–process–outcome model

This model distinguishes:

• Structure: resources and organizational conditions, such as staffing, facilities, and equipment.
• Process: what providers and patients do, such as diagnosis, communication, and treatment.
• Outcome: effects on health, function, satisfaction, and experience.

The categories are analytically helpful but interconnected. Adequate structure does not guarantee good processes, and good processes may not produce desired outcomes when patients have severe disease or social barriers.

Designing an Evaluation: A Practical Sequence

A defensible evaluation can be developed through the following sequence.

1. Clarify the decision. Identify what will be decided, by whom, and when.
2. Describe the intervention and theory. Define its components, target population, mechanism, and expected results.
3. Engage stakeholders. Identify perspectives, information needs, risks, and feasible forms of participation.
4. Specify evaluation questions and criteria. Define what counts as adequate performance or value.
5. Select indicators and methods. Combine measures appropriate to the question; do not select indicators merely because data are easy to obtain.
6. Establish a comparison or baseline where feasible. Comparisons may involve time, sites, groups, standards, or expected performance.
7. Plan for data quality and ethics. Address validity, reliability, missing data, privacy, consent, security, and conflicts of interest.
8. Analyze context and unintended effects. Look beyond intended outcomes and average results.
9. Interpret with stakeholders. Test explanations without allowing opinion to replace evidence.
10. Communicate and use findings. Match reports, dashboards, briefings, and feedback cycles to users and decisions.

Evaluation quality is multidimensional. A highly rigorous design that cannot be implemented is not useful; a feasible design with invalid measures is not credible. The appropriate standard is fitness for purpose: sufficient rigor, transparency, timeliness, and relevance for the decision at hand.

Ethical, Equity, and Governance Considerations

Evaluation must protect participants and communities. Key issues include informed consent where required, confidentiality, secure data handling, minimization of identifiable data, fair participant selection, and management of potential harms. Evaluation involving vulnerable groups requires particular attention to coercion and unequal power.

The distinction between quality improvement, evaluation, and research can affect review requirements, but labels should not be used to avoid ethical oversight. Organizations should follow applicable institutional, legal, and professional requirements.

Equity should be integrated from the start. Evaluation should ask who benefits, who is missed, who bears risks, and whether apparent average improvement conceals deterioration among marginalized groups. Data should be disaggregated where appropriate by factors such as age, sex, disability, ethnicity, geography, socioeconomic position, and language, while avoiding stigmatization and protecting privacy.

Governance also concerns independence. Funders may have legitimate interests, but they should not suppress unfavorable results. Evaluation reports should disclose sponsorship, conflicts of interest, limitations, deviations from the original plan, and uncertainty.

Technology and Artificial Intelligence

Electronic health records, remote monitoring, patient-reported outcome systems, and data linkage expand evaluation capacity. They also introduce risks: coding variation, missingness, interoperability problems, surveillance concerns, and unequal digital access.

Artificial intelligence may support case identification, prediction, text analysis, and rapid synthesis. However, an algorithm’s apparent performance is not equivalent to program impact. Evaluation should examine calibration, subgroup performance, workflow effects, explainability appropriate to the use, automation bias, and consequences of false positives and false negatives. Models can degrade when populations, clinical practice, or data systems change. Human oversight and post-deployment monitoring are therefore essential.

Illustrative Case: Evaluating a Hospital-to-Home Transition Program

Background. A regional hospital introduces a nurse-led transition program for older adults discharged after acute medical admission. The program includes medication reconciliation, a discharge call within 48 hours, a primary-care appointment, and escalation for warning symptoms.

Problem. After six months, readmissions appear lower, but managers do not know whether the reduction is real, whether the program reaches high-risk patients, or whether it creates workload and access problems elsewhere.

Analysis. The evaluation team develops a logic model. Inputs include nurses, a risk-identification tool, pharmacy support, and appointment capacity. Activities include discharge education, medication review, and follow-up calls. Outputs include completed reconciliations and successful contacts. Outcomes include medication discrepancies, patient understanding, timely primary-care review, emergency visits, readmissions, patient experience, and staff workload.

The team uses a mixed-method design. Routine data are compared with the six months before implementation, while a similar hospital without the program provides a supplementary comparison. Interviews with patients and staff explore barriers, adaptations, and unintended effects. Results are stratified by language, age, rural residence, and deprivation where data quality permits.

Approach and interpretation. The program reaches only 62% of eligible patients because weekend discharges and incomplete telephone information create barriers. Among reached patients, medication discrepancies decline and patient understanding improves. Readmissions decline modestly, but the comparison hospital also experiences a decline, limiting causal certainty. Staff report that the program shifts work to primary-care practices, and patients with limited English proficiency benefit less because interpreter access is inconsistent.

Outcome. The organization does not simply label the program a success or failure. It retains the core intervention, improves weekend coverage, funds interpreter access, establishes a primary-care capacity agreement, and continues monitoring readmissions and equity indicators. A later evaluation is planned to assess sustainability and longer-term outcomes.

Learning points. The case demonstrates why outcome data alone are insufficient; implementation and context explain reach and variation. It also shows the value of mixed methods, comparison data, equity analysis, and a decision-oriented evaluation cycle.

Challenges and Limitations

Health care evaluation faces several recurring problems:

• Attribution: outcomes may reflect many simultaneous influences.
• Time lag: important benefits or harms may emerge only after years.
• Measurement limitations: administrative indicators may omit experience, function, quality of life, or informal care.
• Data quality: missing, inconsistent, delayed, or selectively recorded data can bias conclusions.
• Pressure for simple answers: complex results are often reduced to a single score or pass/fail label.
• Implementation variation: an intervention may differ across sites, teams, or patient groups.
• Evaluation burden: data collection can divert time from care and create staff fatigue.
• Change during evaluation: policy, staffing, technology, or population conditions may shift before analysis is complete.
• Publication and reporting bias: favorable results may be more likely to be disseminated.

These limitations do not make evaluation impossible. They require explicit assumptions, triangulation, sensitivity to context, transparent uncertainty, and proportionate methods.

Best Practices for Health Care Managers

Managers commissioning or using evaluations should:

1. Begin with a decision, not with available data.
2. Define success using multiple relevant dimensions, including equity and unintended effects.
3. Involve intended users and affected communities early.
4. Combine routine indicators with qualitative understanding where appropriate.
5. Distinguish implementation failure from theory failure.
6. Use comparison or baseline information when feasible, while acknowledging its limits.
7. Protect independence, privacy, and participant welfare.
8. Report limitations and uncertainty honestly.
9. Plan dissemination and action before data collection begins.
10. Treat evaluation as a learning cycle rather than a one-time inspection.

Chapter Summary

Health care evaluation is the systematic use of evidence, criteria, and contextual understanding to judge and improve interventions, services, policies, and organizations. It supports learning, accountability, decision-making, knowledge development, and equity. Evaluation differs from monitoring, audit, assessment, research, and quality improvement, although these activities can overlap and inform one another.

A logic model connects needs and context with inputs, activities, outputs, outcomes, and impacts. Formative evaluation supports design; process evaluation examines implementation; outcome evaluation examines short- and medium-term changes; impact evaluation examines broader or longer-term effects; and summative evaluation provides an overall judgment. Frameworks such as theory of change, CIPP, realist evaluation, utilization-focused evaluation, and the structure–process–outcome model help organize inquiry.

Effective evaluation is decision-oriented, stakeholder-informed, ethically governed, context-sensitive, and proportionate to the question. It combines methodological rigor with practical use, examines equity and unintended effects, and communicates uncertainty rather than overstating conclusions.

Key Terms

Health care evaluation: Systematic inquiry used to judge value, performance, implementation, or effects in health care.

Monitoring: Routine and continuing observation of indicators and trends.

Audit: Comparison of practice or records with defined standards.

Assessment: Examination of needs, conditions, capacity, feasibility, or risk.

Research: Systematic inquiry intended to develop or contribute to knowledge.

Quality improvement: Structured, action-oriented work to improve performance in a particular setting.

Program evaluation: Systematic examination of a program’s rationale, implementation, results, value, and context.

Logic model: Representation of relationships among context, inputs, activities, outputs, outcomes, and impacts.

Formative evaluation: Evaluation used to improve design or feasibility before or during early implementation.

Process evaluation: Examination of delivery, reach, fidelity, adaptation, acceptability, and feasibility.

Outcome evaluation: Examination of short- or medium-term changes associated with an intervention.

Impact evaluation: Examination of broader, longer-term effects and causal contribution.

Summative evaluation: Overall judgment of an intervention at a defined decision point.

Theory of change: Explanation of how and why an intervention is expected to produce results.

Context: Conditions that influence implementation, results, interpretation, or transferability.

Stakeholder: Person or organization affected by, interested in, or able to influence an evaluation.

Fidelity: Extent to which core intervention components are delivered as intended.

Equity: Fair distribution of benefits, burdens, access, and opportunity in health care.

Utilization: Practical use of evaluation findings by intended decision-makers.

Review Questions

Short Answer Questions

1. Define health care evaluation and identify its three central features.
2. What is the difference between monitoring and audit?
3. Explain the purpose of a logic model.
4. Distinguish formative evaluation from summative evaluation.
5. Why is context important when interpreting evaluation findings?

Long Answer Questions

1. Compare monitoring, audit, assessment, research, quality improvement, and program evaluation. Explain where their purposes may overlap.
2. Discuss the roles of stakeholders in planning, conducting, interpreting, and using a health care evaluation.
3. Explain the theory of change and show how it can expose assumptions in a clinical or public health intervention.
4. Compare process, outcome, and impact evaluation, including the types of decisions each can support.
5. Discuss ethical, equity, governance, and data-quality issues that should be addressed in a major health care evaluation.

Analytical / Application Questions

1. A hospital introduces an electronic sepsis-alert system, but mortality does not change after one year. Design an evaluation that distinguishes implementation problems from problems with the intervention theory.
2. A vaccination program reports high overall coverage but low coverage in several minority communities. What evaluation questions, data sources, and stakeholder perspectives would you use to investigate the finding?
3. A health ministry must decide whether to scale a promising telehealth pilot. Develop a decision-oriented evaluation framework covering reach, implementation, outcomes, equity, cost, sustainability, and contextual conditions.

Multiple Choice Questions with Answers

Question 1

Which statement best defines health care evaluation?

A. Routine counting of service activity without interpretation

B. Systematic inquiry to judge value, performance, implementation, or effects

C. Comparison of clinical records with professional standards only

D. Research conducted exclusively with randomized trials

Correct Answer: B. Systematic inquiry to judge value, performance, implementation, or effects

Question 2

What is the primary emphasis of monitoring?

A. Routine observation of indicators and trends

B. Determining whether a program caused an outcome

C. Testing a new theory for general scientific knowledge

D. Judging compliance with a professional code only

Correct Answer: A. Routine observation of indicators and trends

Question 3

Which evaluation type is most appropriate for testing whether a proposed service is acceptable and feasible before launch?

A. Impact evaluation

B. Summative evaluation

C. Formative evaluation

D. Compliance audit

Correct Answer: C. Formative evaluation

Question 4

Which of the following is a process-evaluation question?

A. Did population mortality decline over ten years?

B. Were intended patients reached and did staff deliver the core components?

C. Is the intervention cost-effective compared with all alternatives?

D. What is the national burden of disease?

Correct Answer: B. Were intended patients reached and did staff deliver the core components?

Question 5

In a logic model, staff training sessions delivered to personnel are usually classified as:

A. Inputs

B. Activities

C. Outputs

D. Impacts

Correct Answer: C. Outputs

Question 6

What is the central question of realist evaluation?

A. Does the intervention meet a documentation standard?

B. What works, for whom, in what circumstances, and how?

C. How many services were delivered last month?

D. Was the intervention approved by a regulator?

Correct Answer: B. What works, for whom, in what circumstances, and how?

Question 7

Which feature most clearly distinguishes an audit from a general program evaluation?

A. An audit compares practice with an explicit standard

B. An audit always uses qualitative interviews

C. An audit examines only long-term population impacts

D. An audit cannot be used for corrective action

Correct Answer: A. An audit compares practice with an explicit standard

Question 8

Why should an evaluation examine equity as well as average outcomes?

A. Average outcomes are never useful

B. Equity analysis can reveal groups that benefit less or experience greater burdens

C. Equity analysis eliminates the need for stakeholder involvement

D. Equity is relevant only to financial evaluations

Correct Answer: B. Equity analysis can reveal groups that benefit less or experience greater burdens

Question 9

A program reaches only half of its intended population, but participants show improved outcomes. Which conclusion is most justified?

A. The program is completely ineffective

B. The program is successful for everyone

C. The program may be effective among participants but has limited reach

D. No further evaluation is needed

Correct Answer: C. The program may be effective among participants but has limited reach

Question 10

Which practice best supports trustworthy evaluation governance?

A. Suppressing unfavorable findings to protect the program

B. Collecting every possible data item regardless of burden

C. Disclosing conflicts of interest, limitations, and deviations from the plan

D. Allowing the funder to change results after analysis

Correct Answer: C. Disclosing conflicts of interest, limitations, and deviations from the plan