Strategic Alignment and Evidence-Based HRM

Introduction

Human resource analytics creates value only when it helps an organization make better decisions about people, work, and organizational capability. A technically sophisticated dashboard is not strategically useful if it measures activity rather than outcomes, answers questions no decision-maker is asking, or encourages action that conflicts with organizational values. Strategic alignment therefore provides the foundation for effective HR analytics.

Strategic alignment means connecting organizational purpose and priorities with workforce decisions, HR practices, evidence, and measurable outcomes. Evidence-based human resource management (EBHRM) extends this idea by requiring decisions to draw on the best available research evidence, internal organizational data, professional expertise, and the perspectives of affected stakeholders.

This chapter develops several complementary perspectives. Logic models and theories of change explain how HR interventions are expected to produce results. Human capital theory explains why employee knowledge and capability may generate economic value. The resource-based view considers when people and HR systems can create sustained advantage. Analytical maturity models help organizations assess their ability to use data responsibly. Together, these frameworks help analysts translate broad business problems into answerable HR questions.

Learning Objectives

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

1. Explain strategic alignment and evidence-based HRM.
2. Use human capital theory and the resource-based view to interpret HR value creation.
3. Construct a logic model and theory of change for an HR intervention.
4. Assess an organization’s analytical maturity.
5. Translate a business problem into specific, answerable HR questions.
6. Identify stakeholder, ethical, data-quality, and causal-inference issues in HR analytics.
7. Recommend an evidence-informed approach to an HR decision.

1. From HR Activity to Organizational Value

Traditional HR reporting often focuses on activity: the number of employees hired, training hours delivered, vacancies filled, or performance reviews completed. These measures may be useful for operational control, but they do not necessarily demonstrate value. Strategic HR analytics asks a different question: how do workforce conditions and HR practices affect outcomes that matter to the organization and its stakeholders?

A useful distinction is between four levels of measurement:

The levels are related but not interchangeable. More training hours do not prove improved capability, and lower turnover does not automatically prove better performance. Analysts must articulate the mechanism that connects one level to the next and identify alternative explanations.

Strategic alignment involves vertical and horizontal connections. Vertical alignment links HR priorities to organizational strategy. For example, a firm competing through rapid innovation may need analytics focused on technical capability, collaboration, experimentation, and retention of scarce expertise. Horizontal alignment concerns consistency among HR practices. Recruitment, onboarding, development, performance management, reward, and work design should reinforce rather than contradict one another.

Alignment is not simply the process of making HR serve financial targets. Public, nonprofit, and responsible business organizations may pursue service quality, equity, safety, sustainability, employee well-being, and community outcomes alongside financial performance. A strategically aligned analytics function makes these priorities visible rather than reducing all value to a single monetary measure.

2. Evidence-Based Human Resource Management

Evidence-based management is a disciplined approach to decision-making that combines multiple sources of evidence rather than relying on intuition, tradition, authority, or a single dataset. In HRM, the four commonly recognized evidence sources are:

1. Best available scientific evidence: findings from credible research, systematic reviews, and well-designed studies.
2. Organizational evidence: internal data about employees, processes, customers, operations, and outcomes.
3. Professional expertise: informed judgment based on HR knowledge and experience.
4. Stakeholder values and perspectives: the interests, experiences, and rights of employees, managers, customers, communities, and other affected groups.

These sources answer different questions. Research may indicate that structured selection methods tend to improve prediction of job performance, while organizational data show whether a particular implementation works in a particular context. Professional expertise helps interpret feasibility, and stakeholder perspectives reveal consequences that numerical measures may miss.

Evidence-based HRM does not mean that every decision must be reduced to a statistical model. It means that decision-makers should make assumptions explicit, assess evidence quality, consider competing explanations, and evaluate consequences. Evidence is not automatically strong because it is quantitative. A biased dataset, poorly defined measure, or correlational association may be less useful than carefully collected qualitative evidence.

Evidence quality

Analysts should examine:

• Validity: Does the evidence measure or support the claim it is being used to make?
• Reliability: Would the measurement produce reasonably consistent results under similar conditions?
• Relevance: Does the evidence apply to this population, job, time period, and decision?
• Transparency: Are definitions, methods, assumptions, and limitations documented?
• Representativeness: Are important groups included, or are some voices systematically absent?
• Timeliness: Is the evidence current enough for the decision?

A practical evidence review should distinguish description, prediction, and explanation. A report describing higher turnover in one department does not establish why turnover is higher. A predictive model estimating who may leave does not establish that a particular intervention will prevent leaving. Causal claims require stronger designs and careful reasoning.

3. Human Capital Theory and the Creation of Value

Human capital theory treats employees’ knowledge, skills, abilities, experience, health, and other productive attributes as forms of capital that can generate future benefits. Individuals and organizations may invest in education, training, job design, health, and development because these investments can increase productive capacity.

The theory supports an important HR analytics question: which workforce capabilities contribute to which organizational outcomes, and under what conditions? For example, technical training may improve productivity only when employees have suitable equipment, supportive supervisors, opportunities to apply learning, and work processes that permit discretion.

A simplified investment logic can be represented as follows:

Benefits may include additional output, better quality, reduced errors, lower absence, improved safety, or stronger innovation. Opportunity costs include time away from productive work and resources that could have been used elsewhere. The expression is a decision framework, not a complete valuation formula. It requires credible estimates, a defined time horizon, and attention to uncertainty.

Human capital is not owned in the same way as physical equipment. Employees can leave, skills can become obsolete, and capability depends on social and organizational conditions. Consequently, analytics should examine not only individual characteristics but also knowledge sharing, team composition, leadership, technology, and work design.

A major limitation of simple human capital reasoning is that it can portray employees primarily as economic inputs. A responsible HR analyst must also consider dignity, autonomy, fairness, well-being, and the possibility that an intervention produces value for one stakeholder while imposing costs on another.

4. The Resource-Based View and Strategic HRM

The resource-based view (RBV) argues that sustained competitive advantage depends on resources and capabilities that are valuable, rare, difficult to imitate, and difficult to substitute. In many interpretations, the relevant resource is not an isolated employee but a system of human and organizational capabilities: specialized expertise, routines, culture, leadership, relationships, and coordinated ways of working.

The VRIO framework is commonly used to examine strategic resources:

Analytics can support each question, but cannot answer all of them through employee-level prediction. For example, a firm may possess rare engineering expertise, but if knowledge is concentrated in a few individuals and not transferred through systems, the capability may be fragile. An analyst should therefore examine succession risk, collaboration networks, documentation, internal mobility, and learning processes.

The RBV also warns against simplistic claims that “people are our greatest asset.” Employees and capabilities create value in context. A capability that is valuable in one competitive strategy may be less relevant in another. Moreover, advantage may not be sustainable if competitors can hire the same talent, acquire the technology, or reproduce the work system.

5. Logic Models and Theories of Change

A logic model is a structured representation of how resources and activities are expected to lead to outputs, outcomes, and impacts. It makes assumptions visible and helps determine what should be measured.

A generic HR logic model has the following sequence:

Inputs → Activities → Outputs → Short-term outcomes → Longer-term outcomes → Impact

For a leadership development program:

• Inputs: budget, facilitators, participant time, learning materials.
• Activities: workshops, coaching, practice assignments, feedback.
• Outputs: attendance, completed assignments, coaching sessions.
• Short-term outcomes: increased knowledge, confidence, and skill.
• Longer-term outcomes: improved supervisory behavior, team climate, and retention.
• Impact: better service quality, productivity, employee well-being, or organizational performance.

A logic model is not proof that the sequence is true. It is a testable proposition. Analysts should identify assumptions such as: participants attend; learning transfers to the job; managers support behavior change; the labor market remains stable; and other factors do not overwhelm the intervention.

A theory of change is usually more explanatory than a logic model. It specifies why and under what conditions change is expected. It may include:

1. The problem and its causes.
2. The desired long-term change.
3. The mechanisms through which activities influence behavior.
4. Contextual conditions and risks.
5. Evidence indicators and evaluation points.

For instance, an onboarding intervention may assume that early role clarity reduces uncertainty, which improves confidence and social integration, which increases performance and reduces early turnover. The theory can be tested by measuring role clarity and integration rather than only counting onboarding sessions.

Measurement implications

Each link requires an appropriate measure. If the proposed mechanism is increased role clarity, a completion rate is insufficient. If the intended impact is customer service, employee survey scores alone are insufficient. A balanced measurement plan may combine administrative records, validated surveys, interviews, manager assessments, customer outcomes, and qualitative observations.

Analysts should also distinguish leading and lagging indicators. Leading indicators, such as access to coaching or clarity of goals, may change before performance outcomes. Lagging indicators, such as turnover or customer complaints, show what has already occurred. Neither is universally superior; their usefulness depends on the decision and time horizon.

6. Translating Business Problems into Answerable HR Questions

Business leaders often present broad concerns: “Productivity is falling,” “We cannot retain talent,” or “Should we introduce hybrid work?” These are starting points, not analytical questions. The analyst’s task is to clarify the decision, population, outcome, time period, and plausible mechanisms.

A useful translation process is:

Step 1: Define the decision

What decision will be made, by whom, and by when? Analytics without a decision context often produces interesting but unusable information.

Step 2: Define the business outcome

Specify the outcome operationally. “Productivity” might mean units per paid hour, cases resolved with acceptable quality, revenue per employee, or project milestones achieved. Different definitions can lead to different conclusions.

Step 3: Identify the HR domain and mechanism

Is the issue related to staffing, capability, leadership, work design, reward, employee relations, health, inclusion, or technology? What mechanism could plausibly connect the HR factor to the outcome?

Step 4: Specify the population and comparison

Define which employees, locations, jobs, or teams are included. Consider comparison groups, pre-intervention periods, or alternative practices.

Step 5: Formulate answerable questions

Good questions are specific and decision-relevant. Examples include:

• Which factors are associated with voluntary turnover among customer-service employees during the first year of employment?
• Did the structured onboarding program improve time to proficiency compared with the previous onboarding process?
• Which staffing patterns are associated with service quality after accounting for demand and case complexity?
• Does flexible work affect performance differently across job families, caregiving groups, or levels of supervisory support?

Step 6: Determine the evidence needed

Identify data sources, measures, analytic methods, ethical constraints, and the level of confidence required. A descriptive question may require a trend analysis; a causal question may require a quasi-experimental design or carefully constructed comparison.

Step 7: Agree on action thresholds

Decision-makers should clarify what result would justify action. For example, an intervention might be expanded if it improves a defined outcome without creating unacceptable inequity, workload, or cost.

A question-quality test

An HR analytics question is stronger when it is:

• Specific: key concepts and population are defined.
• Relevant: it relates to a real decision.
• Answerable: suitable data and methods are available.
• Ethically acceptable: privacy, fairness, and proportionality are respected.
• Actionable: plausible actions follow from the result.
• Causally disciplined: it does not assume that association proves cause.

7. Analytical Maturity in HR

Analytical maturity describes an organization’s capacity to use data and evidence effectively. Maturity is broader than software ownership. It includes data quality, skills, governance, leadership, culture, methods, and the ability to convert insight into responsible action.

A practical five-level model is:

The levels are not a ladder in which prediction is always better. A well-designed descriptive measure may be more valuable than an inaccurate predictive model. Organizations should build foundational data definitions and governance before adopting advanced techniques.

Maturity should be assessed across dimensions such as:

• Data integration and quality.
• Analytical and statistical capability.
• HR and business domain knowledge.
• Leadership sponsorship.
• Decision processes and adoption.
• Privacy, security, and fairness governance.
• Evaluation and continuous learning.

A mature organization also knows when not to automate. Some decisions involve rights, livelihood, or sensitive personal circumstances and require human review, explanation, appeal, and the possibility of correction.

8. Stakeholders, Ethics, and Governance

HR analytics affects people directly. Stakeholder analysis should identify who supplies data, who is analyzed, who makes decisions, who bears risks, and who benefits. Employees may reasonably ask what data are collected, why they are used, who can access them, how long they are retained, and whether automated recommendations affect opportunities.

Core governance principles include:

• Purpose limitation: collect and use data for specified legitimate purposes.
• Data minimization: use only data necessary for the decision.
• Transparency: explain measures, methods, and significant limitations.
• Confidentiality and security: protect identifiable and sensitive information.
• Fairness: assess disparate impacts and avoid unjustified discrimination.
• Accountability: assign responsibility for decisions and errors.
• Human oversight: provide review and appeal for consequential decisions.
• Validity: do not use a measure merely because it is available.

Legal requirements vary by jurisdiction, but ethical responsibility extends beyond minimum compliance. A model can be legally permissible yet unfair, intrusive, or organizationally harmful. Analysts should document data provenance, access controls, retention rules, model performance across relevant groups, and procedures for correcting inaccurate records.

Artificial intelligence can support classification, forecasting, text analysis, and recommendation, but it does not remove the need for strategic reasoning. Historical HR data may reproduce past bias. Proxy variables may encode protected characteristics. A model optimized for predictive accuracy may still be unsuitable if its errors are distributed unfairly or if managers misuse its outputs. Human judgment is not automatically unbiased either; governance should address both automated and human decision processes.

9. Worked Application: Reducing Early-Career Turnover

The following is an illustrative fictional case.

Background

Northstar Services, a 2,000-employee customer-support organization, reports that first-year voluntary turnover has increased. Senior leaders propose a predictive model to identify employees likely to resign.

Problem clarification

The initial request—“build a turnover model”—is reframed as a decision problem: what evidence can guide actions that improve first-year retention without unfairly labeling employees or increasing surveillance?

The team defines first-year voluntary turnover as an employee-initiated separation occurring within 365 days of hire. It also identifies a business concern: turnover creates recruitment and training costs and disrupts service continuity. The team agrees that retention must not be pursued through coercive practices or by restricting employee mobility.

Theory of change

Northstar hypothesizes that unclear expectations, inconsistent supervisor support, weak early social integration, and unpredictable scheduling increase early frustration. Improved realistic job information, structured onboarding, supervisor check-ins, schedule predictability, and career conversations may improve role clarity and belonging, thereby reducing avoidable turnover.

Evidence plan

The team combines:

• HR records: hire date, separation date, job, location, supervisor, and schedule changes.
• Employee survey data: role clarity, supervisor support, workload, and belonging.
• Interviews with recent leavers and stayers.
• Operational data: training completion, absence, quality, and service demand.
• Research evidence on realistic job previews, onboarding, supervisory support, and turnover.

The analysis begins descriptively. Turnover rates are compared across job groups and cohorts, with confidence intervals and missing-data checks. Associations are then examined while recognizing that association does not establish cause. If an intervention is introduced in selected sites at different times, the organization may compare changes over time between participating and not-yet-participating sites, provided the design assumptions are examined.

Intervention

Northstar introduces a realistic job preview before hiring, a 90-day onboarding pathway, supervisor check-ins at weeks 2, 6, and 12, and improved schedule-notice standards. The organization evaluates both retention and possible unintended effects, including supervisor workload, employee stress, customer outcomes, and differential effects across groups.

Outcome assessment

The team tracks:

• First-year voluntary turnover.
• Time to proficiency.
• Quality and customer outcomes.
• Role clarity and belonging.
• Schedule predictability.
• Employee perceptions of fairness.
• Effects by job family, location, and relevant demographic groups where lawful and appropriate.

A predictive model may still be explored for resource planning, but it is not used to deny training, promotion, or scheduling opportunities. If used, it is subject to validation, documentation, access controls, bias testing, and human review.

Learning points

The case illustrates that strategic alignment precedes model selection. The important issue was not whether turnover could be predicted, but whether the organization understood plausible causes and could take fair, effective action. A theory of change guided measurement; stakeholder analysis shaped governance; and maturity considerations prevented the organization from treating prediction as a substitute for intervention design.

10. Professional Practice and Implementation Guidance

HR analysts can improve strategic value through several practices.

First, begin with the decision rather than the dataset. Data availability should not determine the organization’s most important questions. Second, define measures in a shared data dictionary. Terms such as “headcount,” “turnover,” “absence,” and “performance” often have multiple operational meanings. Third, combine quantitative and qualitative evidence. Numbers can show where a pattern occurs, while interviews and observation may clarify how it is experienced.

Fourth, use the least complex method that adequately answers the question. Descriptive analysis, segmentation, process mapping, and careful comparison may be sufficient. More advanced models require stronger technical capability and governance. Fifth, distinguish prediction from causation and communicate uncertainty. Decision-makers should know what is known, what is inferred, and what remains uncertain.

Sixth, evaluate interventions rather than merely reporting them. Establish baseline measures, define comparison strategies where feasible, monitor implementation fidelity, and look for unintended consequences. Finally, close the evidence-to-action loop: record the decision, monitor results, review stakeholder feedback, and revise the theory of change when evidence contradicts assumptions.

Challenges and Limitations

Strategic HR analytics faces several persistent challenges. Organizational outcomes are influenced by many simultaneous factors, making causal attribution difficult. Employee data may be incomplete, inconsistent, or collected for administrative rather than analytical purposes. Small groups may create privacy risks or unstable estimates. Survey responses may reflect fear, low trust, or nonresponse bias. Measures can also create perverse incentives: if managers are rewarded for reducing absence, they may discourage legitimate reporting.

Strategic alignment may become narrow if leaders define value only through short-term financial results. Conversely, an overly broad set of objectives can make prioritization impossible. Frameworks such as human capital theory and the RBV are useful lenses, but neither provides a complete account of power, identity, employee voice, institutional context, or ethical responsibility. The analyst must therefore use frameworks critically, not mechanically.

Contemporary Developments

HR analytics is increasingly integrated with workforce planning, skills intelligence, employee listening, internal talent marketplaces, and responsible AI governance. Organizations are moving from static job descriptions toward skills-based approaches, although skill taxonomies can be incomplete or biased. Real-time people data may support faster decisions, but speed increases the risk of intrusive monitoring and reaction without reflection.

The growing use of generative AI also raises questions about explainability, confidentiality, hallucinated outputs, intellectual property, and accountability. AI tools can assist with summarizing qualitative data or drafting reports, but analysts must verify outputs, protect sensitive information, and retain responsibility for interpretation. The strategic principle remains unchanged: technology should serve a legitimate decision and a defensible theory of value.

Chapter Summary

Strategic HR analytics connects organizational priorities, workforce conditions, HR practices, evidence, and stakeholder value. Evidence-based HRM combines scientific research, organizational data, professional expertise, and stakeholder perspectives. Human capital theory explains how investments in knowledge and capability may produce benefits, while the resource-based view examines whether human and organizational capabilities are valuable, rare, difficult to imitate, and supported by the organization.

Logic models map inputs, activities, outputs, outcomes, and impacts. Theories of change add causal mechanisms, assumptions, context, and risks. These tools help analysts select meaningful measures rather than relying on convenient activity metrics. Analytical maturity develops from ad hoc reporting toward descriptive, diagnostic, predictive, and strategically embedded decision support, but advanced prediction is not automatically superior.

Effective analysts translate vague business concerns into specific questions about decisions, populations, outcomes, mechanisms, comparisons, evidence, and action thresholds. They distinguish association from causation, communicate uncertainty, protect privacy, examine fairness, and evaluate unintended consequences. The central professional responsibility is to ensure that analytics improves decisions while respecting the dignity, rights, and interests of the people represented in the data.

Key Terms

Strategic alignment: Connection between organizational strategy, HR practices, workforce capabilities, and measures of value.

Evidence-based HRM: HR decision-making informed by research evidence, organizational data, professional expertise, and stakeholder perspectives.

Human capital: Knowledge, skills, abilities, experience, health, and other employee attributes that may generate future value.

Resource-based view: A perspective that explains advantage through valuable, rare, difficult-to-imitate, and well-organized resources and capabilities.

VRIO: Framework assessing whether a resource is Valuable, Rare, Inimitable, and supported by Organization.

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

Theory of change: An explanation of how and why an intervention is expected to produce specified changes under particular conditions.

Input: Resource required to undertake an activity.

Output: Immediate product or result of an activity.

Outcome: Change in behavior, capability, condition, or performance following an intervention.

Impact: Longer-term organizational or stakeholder consequence.

Leading indicator: Measure that may signal future performance or outcomes.

Lagging indicator: Measure reflecting an outcome that has already occurred.

Analytical maturity: Organizational capability to produce, govern, interpret, and use data and evidence.

Descriptive analytics: Analysis of what has happened.

Diagnostic analytics: Analysis intended to understand patterns and possible contributing factors.

Predictive analytics: Estimation of likely future outcomes or events.

Causal inference: Reasoning about whether an intervention or exposure produced an outcome.

Stakeholder: Person or group affected by, contributing to, or responsible for a decision.

Data minimization: Limiting data collection and use to what is necessary for a legitimate purpose.

Review Questions

Short Answer Questions

1. Distinguish between an HR output, outcome, and impact.
2. Name the four principal evidence sources in evidence-based HRM.
3. What does the “O” in the VRIO framework represent?
4. Why is a logic model not proof of causation?
5. State two principles of responsible HR analytics governance.

Long Answer Questions

1. Explain how human capital theory and the resource-based view differ and how they can complement one another in HR analytics.
2. Describe how a theory of change can improve the evaluation of a leadership development program.
3. Discuss the five levels of analytical maturity and explain why predictive analytics is not always the best choice.
4. Explain the process of translating “employee turnover is too high” into an answerable HR analytics question.
5. Evaluate the role of stakeholder perspectives in evidence-based HRM.

Analytical / Application Questions

1. A hospital wants to reduce nurse turnover. Construct a logic model identifying inputs, activities, outputs, outcomes, impacts, and at least three assumptions.
2. A company uses a predictive model to identify employees likely to resign and then excludes them from development programs. Analyze the strategic, ethical, and methodological problems in this practice and recommend an alternative.
3. An organization reports that teams with more training hours have higher productivity. Explain at least three alternative explanations and propose a stronger evaluation design.

Multiple Choice Questions with Answers

Question 1

Which statement best describes evidence-based HRM?

A. Using only employee survey data to make HR decisions

B. Combining research evidence, organizational data, professional expertise, and stakeholder perspectives

C. Allowing senior managers to make decisions without formal analysis

D. Replacing professional judgment with predictive algorithms

Correct Answer: B. Combining research evidence, organizational data, professional expertise, and stakeholder perspectives

Question 2

In a logic model, which is an outcome rather than an output?

A. Number of employees attending training

B. Number of coaching sessions delivered

C. Improved supervisory behavior after training

D. Training budget approved

Correct Answer: C. Improved supervisory behavior after training

Question 3

Which question is most clearly strategic and answerable?

A. Why are employees unhappy?

B. Can HR improve everything?

C. Which factors are associated with first-year voluntary turnover among customer-service employees?

D. Do managers need to do better?

Correct Answer: C. Which factors are associated with first-year voluntary turnover among customer-service employees?

Question 4

In the VRIO framework, a resource that is valuable but common among competitors is unlikely by itself to provide what?

A. Any operational usefulness

B. A sustained competitive advantage

C. Any employee benefit

D. A measurable output

Correct Answer: B. A sustained competitive advantage

Question 5

Which analytical maturity level focuses primarily on understanding why a pattern occurred?

A. Ad hoc

B. Descriptive

C. Diagnostic

D. Predictive

Correct Answer: C. Diagnostic

Question 6

Which statement correctly distinguishes prediction from causation?

A. A predictive association always proves an intervention caused the outcome

B. Prediction estimates likely outcomes, whereas causal analysis examines whether an intervention produced an outcome

C. Causation can be established from any correlation

D. Prediction and causation are identical terms

Correct Answer: B. Prediction estimates likely outcomes, whereas causal analysis examines whether an intervention produced an outcome

Question 7

Which is an example of data minimization?

A. Collecting every available employee characteristic

B. Keeping personal data indefinitely in case it becomes useful

C. Using only the employee information necessary for a defined legitimate purpose

D. Sharing raw HR data with all managers

Correct Answer: C. Using only the employee information necessary for a defined legitimate purpose

Question 8

A company finds that employees who receive more training have higher performance. Which is a plausible alternative explanation?

A. Training can never affect performance

B. High-performing employees may be more likely to be selected for training

C. Performance cannot be measured

D. All training data are necessarily false

Correct Answer: B. High-performing employees may be more likely to be selected for training

Question 9

Which evidence source is most directly concerned with the experiences and interests of affected employees?

A. Stakeholder perspectives

B. Algorithmic accuracy only

C. Accounting depreciation

D. Industry market share

Correct Answer: A. Stakeholder perspectives

Question 10

Why should a predictive turnover model not automatically be used to deny development opportunities?

A. Prediction models never produce useful information

B. The model may be inaccurate or biased, and denying opportunities can create unfair and self-fulfilling outcomes

C. Development programs have no relationship to retention

D. Employees cannot change jobs after training

Correct Answer: B. The model may be inaccurate or biased, and denying opportunities can create unfair and self-fulfilling outcomes