The AMO Framework: How Ability, Motivation, and Opportunity Influence Employee Performance

The AMO framework states that an employee's performance depends on their ability to do the work, their motivation to do it, and the opportunity their work situation gives them to do it. In human resource research it has become the most common way of explaining how HR practices might affect performance: selection and training are said to act on ability, pay and appraisal on motivation, and job design and employee participation on opportunity. For a manager its value is diagnostic, because the remedies are not interchangeable. Retraining someone blocked by a broken approval process wastes the training budget, and a bonus offered to someone who lacks the skill rewards nothing.

The textbook summary compresses three things this article opens up: a history with several sources rather than one, a meaning of opportunity that changed on its way into HR, and a formula, P = f(A × M × O), whose claim that performance collapses when any component is missing remains a genuine methodological dispute. AMO is also the usual theoretical lens for research on high-performance work systems, whose evidence is the subject of Article 19, and the motivation theories of Maslow, Herzberg and McClelland are evaluated in Article 23.

Early industrial psychology mostly explained performance through who was hired and how they were trained, while social psychologists put more weight on motivation (Kellner, Cafferkey and Townsend, 2019). The two traditions were joined in a short rule. Marvin Dunnette, reviewing the field in 1973, recalled that after Brayfield and Crockett's 1955 review found no stable relationship between job satisfaction and performance, industrial psychologists were "spurred on by Maier's dictum that Performance is a function of Ability multiplied by Motivation" (Dunnette, 1973, p. 4). The dictum is usually traced to Norman Maier's textbook Psychology in Industry (1955). Victor Vroom's expectancy theory then gave it a theoretical basis: in Work and Motivation he wrote that "the effects of ability and motivation on performance are not additive but interactive" (Vroom, 1964, p. 203, as quoted by Van Iddekinge et al., 2018).

The multiplication was a theoretical claim, not a measured result, though its intuition is strong: an able person who makes no effort produces nothing. Doubts appeared early. Dunnette's report summarized experiments in which ability accounted for far more of the variation in performance than motivational measures such as expectancy and instrumentality. Until psychologists could specify the motivation term in "Maier's equation" with more certainty, he concluded, they "had best leave an unknown term (What?) there and seek to discover its nature by continued empirical studies" (p. 12). The title of his report put the question plainly: Performance Equals Ability and What? The opportunity term is one answer to it.

Ability multiplied by motivation locates performance entirely inside the person. Organizational researchers increasingly objected that this left out the work itself. Kellner and colleagues (2019) found the earliest reference to an opportunity element in a brief passage of Campbell, Dunnette, Lawler and Weick's Managerial Behavior, Performance, and Effectiveness (1970). In 1980 Lawrence Peters and Edward O'Connor argued that situational constraints on work were, in the words of their title, "a frequently overlooked construct" in explaining work outcomes.

The fullest statement came two years later from Melvin Blumberg and Charles Pringle in the Academy of Management Review. They argued that the ability-motivation model was "unable to account for environmental variables not under the control of the individual" (Blumberg and Pringle, 1982, p. 563, quoted in Kellner et al., 2019), and they proposed three determinants under different names: capacity, willingness and opportunity. They defined opportunity as "the field of forces surrounding a person and his or her tasks that enables or constrains that person's task performance" (p. 565, quoted in Bos-Nehles et al., 2023), and grouped those forces as "states of nature" and "actions of others" (p. 563).

The component lists summarize Blumberg and Pringle's examples as reported by Marin-Garcia and Martinez Tomas (2016) and Kellner et al. (2019); the original article was not available in full for this series.

For performance theory, the credit for adding opportunity therefore belongs to Blumberg and Pringle (1982), with Campbell and colleagues (1970) as a brief precursor and Peters and O'Connor (1980) approaching the same problem through constraints. HR research acquired the term by a different route.

The HR version is generally dated to Thomas Bailey's 1993 working paper for Columbia University, Discretionary Effort and the Organization of Work: Employee Participation and Work Reform Since Hawthorne. As later authors report it (the paper itself was not available for this article), Bailey argued that discretionary effort required workers with the necessary skills, appropriate motivation, and an opportunity to participate (Marin-Garcia and Martinez Tomas, 2016, summarizing Appelbaum et al., 2000). Bailey then co-wrote Manufacturing Advantage with Eileen Appelbaum, Peter Berg and Arne Kalleberg (2000), based on 44 plants and surveys of more than 4,000 employees in steel, apparel, and medical electronics and imaging, which concluded that companies are more successful "when managers share knowledge and power with workers and when workers assume increased responsibility and discretion."

British research consolidated the framework. Peter Boxall (2003, p. 6) described the logic in its now standard form: "performance is seen as a function of employee ability (A), motivation (M) and opportunity to participate or contribute (O)." John Purcell and colleagues placed AMO at the centre of the "People and Performance" model from a CIPD-funded study of large UK organizations, among them Tesco, Nationwide and Selfridges (Purcell et al., 2003), and showed that front-line managers largely determine how employees experience HR practices (Purcell and Hutchinson, 2007). The CIPD's 2025 evidence review gives the same lineage: AMO entered HR "primarily through the work of Bailey (1993) and later Appelbaum et al (2000)."

Two changes happened on the way, and both matter. The first is a change of level. Before about 2000 the framework was applied mainly to individuals; in HR research it now mostly describes bundles of practices said to enhance individual characteristics (Bos-Nehles et al., 2023). Kellner and colleagues (2019) observe that Appelbaum et al. became "the standard reference for the AMO framework in the HRM discipline" while omitting its earlier lineage, and that in the shift to HR systems "the individual became somewhat lost." The second is a change in what opportunity means.

The comparison is this article's synthesis of the sources cited, not a table from either source.

The narrowing matters. An employee can sit on a problem-solving team and still lack a working machine or the information needed to act. HR studies that measure only participation can miss the commonest operational causes of poor performance. Some recent articles even trace AMO to marketing research of the early 1990s, overlooking psychological roots decades older (Kellner et al., 2019). AMO has no single origin; it is a convergence of several lines of work that used similar words for somewhat different things.

Blumberg and Pringle defined ability narrowly, as the "physiological and cognitive capabilities that enable an individual to perform a task effectively" (p. 563, quoted in Bos-Nehles et al., 2023). HR research usually works with a broader definition: the knowledge, skills and competencies a particular job requires. Two families of practice act on it, through different mechanisms.

Selection changes ability by changing who holds the job. Its effect depends on how well the method predicts performance and how much choice the organization has among applicants, and recent evidence has lowered expectations of the first. Re-analysing the meta-analytic record, Paul Sackett, Charlene Zhang, Christopher Berry and Filip Lievens (2022) concluded that earlier work had overcorrected for range restriction. Their estimate of the validity of cognitive ability tests was .31, against the .51 widely cited from Schmidt and Hunter (1998), and structured interviews, at .42, ranked highest among the procedures they examined. Selection still matters, but hiring predicts ability less precisely than a generation of HR textbooks implied, and its gains appear only among people hired after the method improves. Module 08 covers selection in detail.

Training changes ability among people already employed, and it carries a dependency that the ability-motivation formula hides. A skill learned in a classroom improves performance only if the employee has occasion to use it and wants to, which is why the same programme can work in one unit and vanish without trace in another. Module 09 covers training design. The diagnostic point is that when an experienced employee underperforms, more training is the remedy most often proposed and frequently the least likely to help.

Ability has the strongest empirical claim of the three. In Van Iddekinge and colleagues' (2018) meta-analysis the corrected correlation with performance was .44 for cognitive ability and .29 for motivation. Among 174 line managers studied by Anna Bos-Nehles, Maarten van Riemsdijk and Kees Looise (2013), ability was the only component with a significant direct effect on how well managers implemented HR practices, as rated by 1,065 of their subordinates.

AMO does not contain a theory of motivation. It borrows one, and the choice matters more than the acronym suggests. The multiplicative formula takes its theoretical justification from expectancy theory, in which effort depends on whether people believe effort will produce performance and performance will produce outcomes they value. Content theories, which ask what people need from work, belong to a different tradition (Article 23), and goal-setting, equity and self-determination theories point to different levers again. A study that says it uses "the AMO framework" has told the reader little about what it assumes motivates people.

HR research usually measures motivation through practices such as performance-related pay, appraisal, internal promotion and job security. Their shared mechanism is that they change what an employee can expect to follow from extra effort, acting mainly on discretionary effort, the part of effort an employer can neither specify in a contract nor enforce by supervision (Appelbaum et al., 2000; Boxall and Purcell, 2003).

Two cautions follow. Motivation-enhancing practices motivate whatever they reward: as Steven Kerr's "folly of rewarding A, while hoping for B" (Article 10) implies, an incentive can raise motivation and lower performance at once. And the boundary with opportunity is less clear than the framework's three boxes suggest. Hackman and Oldham's (1976) job characteristics model, tested on 658 employees in 62 jobs across seven organizations, treated autonomy and skill variety as sources of internal motivation, yet AMO-based HR studies usually file the same job design features under opportunity (Boon, Den Hartog and Lepak, 2019). Jiang, Lepak, Hu and Baer's (2012) meta-analysis found that opportunity-enhancing practices, like motivation-enhancing practices, were more positively related to employee motivation than skill-enhancing practices were. If wider discretion works mainly because people care more about work they control, the three components are not independent channels, and a diagnosis that separates them will sometimes credit the wrong one.

In the broad sense, opportunity is whatever in the work situation lets ability and motivation become performance: tools, materials, timely information, time, workable procedures, the authority the job requires, and supervisors and co-workers who help rather than obstruct. In the narrower HR sense it is the chance to participate and use discretion. Both tend to be checked last. The four reasons below are this article's interpretation of the sources cited, not a tested model.

• It has no single owner. HR departments design selection, training and pay systems; opportunity is mostly created by line managers, operations, IT and budget decisions. Kellner and colleagues (2019) put it directly: "Individuals do not 'own' opportunities, nor are they inherent in the HR system." Because front-line managers turn intended practices into experienced ones (Purcell and Hutchinson, 2007), an opportunity problem is often a problem with a manager's own decisions.
• Managers explain through people. Green and Mitchell (1979) modelled how leaders attribute a subordinate's poor performance to the person or the circumstances and respond accordingly. A personal explanation points to training or discipline; a situational one points back at the manager who designed the situation.
• It is hard to measure cleanly. In their review of 495 studies of HR systems, Boon, Den Hartog and Lepak (2019) found the confusion of practices with outcomes "particularly prevalent" in the opportunity bundle, where measures of whether employees feel empowered or experience work pressure were counted as practices.
• It works through the other two. Bos-Nehles, Townsend, Cafferkey and Trullen (2023), reviewing 104 quantitative AMO studies, concluded that "opportunity appears to be a boundary condition": it shapes whether ability and motivation turn into performance, but it "never acts as a stand-alone mediator." Such a component is easy to omit from a model and costly to omit from a diagnosis.

The result is an asymmetry. Ability and motivation problems usually have remedies HR can deliver; opportunity problems often have remedies someone else must pay for or concede, such as a machine, more staff, a changed approval rule or delegated authority. That is a political difficulty as much as an analytical one.

Writing AMO as P = f(A × M × O) is a claim with practical implications, and it competes with at least three alternatives.

The evidence against strict multiplication. The largest test concerns two of the terms. Van Iddekinge, Aguinis, Mackey and DeOrtentiis (2018) obtained raw data from 56 samples to test whether cognitive ability and motivation interact. Across all their samples, the additive effects of ability and motivation accounted for about 91 percent of the explained variance and the interaction for about 9 percent, and the authors concluded that the evidence "does not provide strong or consistent support for the hypothesis that Performance = f(Ability × Motivation)." Two decades earlier, Sackett, Gruys and Ellingson (1998) had found that interactions between ability and personality-based measures of motivation were "not detected at above chance levels." For the three-term model, Marin-Garcia and Martinez Tomas (2016) found only four studies in their review that tested multiplication at all, and Boon and colleagues (2019) found that 69 percent of 516 HR systems were measured with additive indices, with only six studies combining AMO-type sub-bundles non-additively.

The evidence for something other than addition. The additive model does not win by default. Bos-Nehles and colleagues (2013) rejected both the purely additive and the purely multiplicative forms; in their data the effect took the form P = fA(1 + O), with opportunity amplifying ability and motivation adding little. Siemsen, Roth and Balasubramanian (2008) found that a constraining-factor model of knowledge sharing "fits the data better than the traditional multiplicative model." Dasí and colleagues (2021), comparing three forms across 285 projects, found the multiplicative model fitted best overall, and most clearly in advanced, more complex projects. Reviewers even read the same evidence differently: Bos-Nehles and colleagues (2023) state that the three-way interactions they found "confirm that AMO is a multiplicative model," while Marin-Garcia and Martinez Tomas (2016) report that one of those studies (Bello-Pintado, 2015) found no three-way synergy.

Why the question is hard to settle. Absence of evidence for interaction is weak evidence against it. McClelland and Judd (1993) showed that field tests of interaction often have less than 20 percent of the statistical efficiency of well-designed experiments, largely because few people in real workplaces sit at the extremes where interactions show most clearly. Organizations rarely keep employees with near-zero ability or motivation for long, removing the very cases a multiplicative model is about. The interaction's share of explained variance was 17.0 percent for job performance and 10.3 percent for training performance, but 2.3 percent in laboratory tasks (Van Iddekinge et al., 2018), so whatever multiplication exists may be stronger at work than in controlled settings.

The defensible conclusion is limited. Strict multiplication is not established and is probably too strong; pure addition is not established either. The evidence fits mainly additive effects with context-dependent interactions, and a real possibility that a severely deficient component acts as a constraint. For managers this matters in two places. In hiring, Van Iddekinge and colleagues drew the implication that applicants "should be allowed to compensate for lower scores on ability measures with higher scores on motivation measures and vice versa." In diagnosis, the constraint model is the most useful working assumption: it directs attention to the condition that is most severely lacking, which is often the one nobody has examined.

Bangladesh provides a documented case in which ability is plainly present and performance is plainly absent. The figures below are real and dated; the AMO reading of them is this article's.

In unannounced visits to 150 public health facilities in 60 randomly selected rural upazilas between mid-March and mid-April 2002, Nazmul Chaudhury and Jeffrey Hammer (2004) found that 35 percent of health providers were absent. Among physicians the rate was 40 percent at upazila health complexes and "an astounding 74 percent" at upgraded union-level centres staffed by a single doctor; 41 percent of physician posts in the sampled facilities were vacant. These were qualified doctors, so ability was not the obvious constraint, and the authors' evidence pointed elsewhere. Because public providers "are paid on salary, there is no monetary incentive to go to work in the morning," and, as they remarked in a footnote, there was not "much shame or probability of formal sanctions associated with being absent." Place of residence mattered most: among upazila health complex physicians, 36 percent of those living outside the facility's town were absent all day, against 13 percent of those living in it. Having a road within a kilometre of the facility was also strongly associated with doctors being present for at least part of the day.

Later evidence describes the same structure. In a 2019 survey of 308 junior doctors with rural experience, published by Islam and colleagues (2024), 96 percent reported challenges in their last rural posting, and 63 percent had not lived in the accommodation provided, mainly because of insecurity (77 percent) and a lack of basic amenities (72 percent). Doctors with connections to influential local people were more likely to complete their rural service without interruption. A policy analysis by Joarder and colleagues (2018) found that the requirement for new government doctors to serve at least two years in rural areas was "often waived for political affiliates."

Read through AMO, the evidence points to interacting deficits rather than one missing component. Opportunity in Blumberg and Pringle's broad sense (housing, security, roads, the practical possibility of living near the post) was weak, and so was motivation to attend: a fixed salary, rare sanctions, and private practice and postgraduate training pulling doctors toward towns (Chaudhury and Hammer, 2004; Islam et al., 2024). Remedies aimed at a single component would therefore be expected to disappoint. Monitoring attendance acts on motivation but leaves housing and security unchanged, and training would add ability to people who already have it. The 2002 data are correlational and dated, and the authors themselves warned against over-reading some correlates, treating village electrification as a likely marker of wealth or community pressure since "facilities have their own generators." The case shows how the framework organizes evidence, not that any cause has been proved.

Before choosing a remedy, check that the problem is one of employee performance at all, since arrears, sales or complaints can move for reasons no employee controls; then work through the questions below. The questions and the two short scenarios after them are original teaching material, not a validated instrument.

Scenario A (constructed example). Six months ago, a Dhaka-based private bank trained its 40 relationship officers for small businesses in a cash-flow-based method of credit appraisal. Assessments showed most could apply it, and the bonus now rewards loans appraised this way, yet only about one application in five uses it. Officers say the head-office credit committee still asks for collateral papers in the old format, and appraisals done the new way come back marked "additional documents required."

Scenario B (constructed example). A software company in Chattogram moved its customer systems to a new cloud platform. Support engineers' resolution times worsened by about a third, although the team's engagement scores remain the company's highest, staffing and tools are unchanged, and engineers can close tickets themselves. Nearly all the slow tickets involve configuration tasks that did not exist on the old platform, and engineers spend long periods asking two colleagues who joined from the platform vendor.

This case is fictional and was constructed for teaching. The organization, people and figures are invented.

Dhanshiri Rural Credit Foundation is a microfinance institution with 58 branches in Rangpur division. Over the past year, the share of loans overdue by more than 30 days has risen from 3 to 9 percent in 14 of its branches, while the other 44 have stayed near 3 percent. The deputy director for operations tells the board that officers there "have lost their discipline" and proposes a monthly bonus for collection rates above 98 percent.

Before the board votes, the HR manager assembles five facts. A hiring freeze has raised the average caseload in the 14 branches from about 320 to about 450 borrowers per officer. Officers are paid an incentive for each new loan disbursed, and nothing for portfolio quality. A tablet-based appraisal app was introduced last year with a one-day orientation, but audits show most appraisals in the app are completed correctly. In the 14 branches, branch managers approved 22 percent of the loans that officers had flagged as risky, against 6 percent elsewhere. And nine of the 14 branches serve riverbank unions where erosion displaced a number of borrowers last season.

What the exercise is designed to show. The deputy director treats an outcome (arrears) as a measure of behaviour and assumes one missing component. A disciplined diagnosis asks first whether employees control the outcome, then tests each component, and here finds the binding conditions in opportunity (caseload and overrides) and in the direction of motivation, not its absence.

The AMO framework is more useful than its formula. Its history is a long attempt to answer Dunnette's question about what else performance depends on. Blumberg and Pringle pointed at the work situation; the HR literature narrowed that answer to participation and attached it to bundles of practices, which made AMO easier to study and drew it further from the employee whose performance it set out to explain.

Three questions remain open. Functional form is unsettled rather than decided, because field data are poorly suited to detecting interactions. Level is unresolved: measuring practices is not measuring the ability, motivation and opportunity employees have, yet just under 80 percent of AMO variables in Bos-Nehles and colleagues' (2023) review referred to practices. And ownership is awkward, because opportunity is the component most often missing and least often controlled by HR, so applying the framework seriously means raising questions about workloads, approval rules and supervisors' decisions that other managers may prefer not to hear. Article 19 examines what happens when organizations try to build all three components into a high-performance work system. The question to carry forward is simpler: when performance falls short, which of the three conditions has anyone actually checked?

Select a real, persistent performance problem in an organization you know: a team that misses targets, a role with high error rates, a unit whose service quality lags. You may disguise the organization.

Prepare a report of 2,500 to 3,000 words that:

1. defines the performance problem in behavioural terms, separating what employees do from outcomes that also depend on factors outside their control;
2. assesses the evidence for each of ability, motivation and opportunity, using both meanings of opportunity discussed in this article, and states where evidence is missing;
3. argues which functional form (additive, multiplicative, constraining-factor or combined) best fits your case, and explains what that choice implies for the order and cost of remedies;
4. evaluates the remedy currently favoured by management, identifying which component it targets and whether the evidence supports that choice;
5. recommends an intervention plan, naming who in the organization owns each element (HR, line management, operations or others), the likely resistance, and how you would know within six months whether it is working.

Use at least five academic sources, including at least three from this article's reference list. Distinguish your evidence from your inferences, and do not present a correlation in your data as proof of a cause.

Assessment rubric (100 marks)

1. Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. (2000). Manufacturing advantage: Why high-performance work systems pay off. Cornell University Press (ILR Press). Study scope as described on co-author's page: https://arnekalleberg.org/books/manufacturing-advantage/
2. Blumberg, M., & Pringle, C. D. (1982). The missing opportunity in organizational research: Some implications for a theory of work performance. Academy of Management Review, 7(4), 560–569. https://doi.org/10.2307/257222
3. Boon, C., Den Hartog, D. N., & Lepak, D. P. (2019). A systematic review of human resource management systems and their measurement. Journal of Management, 45(6), 2498–2537. https://doi.org/10.1177/0149206318818718
4. Bos-Nehles, A. C., Van Riemsdijk, M. J., & Looise, J. K. (2013). Employee perceptions of line management performance: Applying the AMO theory to explain the effectiveness of line managers' HRM implementation. Human Resource Management, 52(6), 861–877. https://doi.org/10.1002/hrm.21578
5. Bos-Nehles, A., Townsend, K., Cafferkey, K., & Trullen, J. (2023). Examining the ability, motivation and opportunity (AMO) framework in HRM research: Conceptualization, measurement and interactions. International Journal of Management Reviews, 25(4), 725–739. https://doi.org/10.1111/ijmr.12332
6. Chaudhury, N., & Hammer, J. S. (2004). Ghost doctors: Absenteeism in rural Bangladeshi health facilities. World Bank Economic Review, 18(3), 423–441. https://doi.org/10.1093/wber/lhh047
7. Dunnette, M. D. (1973). Performance equals ability and what? (Technical Report 4009). University of Minnesota, Center for the Study of Organizational Performance and Human Effectiveness. ERIC ED085636: https://files.eric.ed.gov/fulltext/ED085636.pdf
8. Kellner, A., Cafferkey, K., & Townsend, K. (2019). Ability, motivation and opportunity theory: A formula for employee performance? In K. Townsend, K. Cafferkey, A. McDermott, & T. Dundon (Eds.), Elgar introduction to theories of human resources and employment relations (pp. 311–323). Edward Elgar. https://doi.org/10.4337/9781786439017.00029 (author manuscript: http://hdl.handle.net/10072/395836)
9. Marin-Garcia, J. A., & Martinez Tomas, J. (2016). Deconstructing AMO framework: A systematic review. Intangible Capital, 12(4), 1040–1087. https://doi.org/10.3926/ic.838
10. Van Iddekinge, C. H., Aguinis, H., Mackey, J. D., & DeOrtentiis, P. S. (2018). A meta-analysis of the interactive, additive, and relative effects of cognitive ability and motivation on performance. Journal of Management, 44(1), 249–279. https://doi.org/10.1177/0149206317702220

Other sources cited

• Bailey, T. (1993). Discretionary effort and the organization of work: Employee participation and work reform since Hawthorne [Working paper]. Teachers College and Conservation of Human Resources, Columbia University.
• Boxall, P. (2003). HR strategy and competitive advantage in the service sector. Human Resource Management Journal, 13(3), 5–20. https://doi.org/10.1111/j.1748-8583.2003.tb00095.x
• Boxall, P., & Purcell, J. (2003). Strategy and human resource management. Palgrave Macmillan. (5th ed., Bloomsbury Academic, 2022; see ch. 7, section "The AMO model of performance," p. 180.)
• Campbell, J. P., Dunnette, M. D., Lawler, E. E., & Weick, K. E. (1970). Managerial behavior, performance, and effectiveness. McGraw-Hill.
• Chaudhury, N., Hammer, J., Kremer, M., Muralidharan, K., & Rogers, F. H. (2006). Missing in action: Teacher and health worker absence in developing countries. Journal of Economic Perspectives, 20(1), 91–116. https://doi.org/10.1257/089533006776526058
• CIPD. (2025). Impact of human resource management: An evidence review. Scientific summary. Chartered Institute of Personnel and Development. https://www.cipd.org/globalassets/media/knowledge/knowledge-hub/evidence-reviews/2025-pdfs/8959-impact-of-hrm-evidence-review.pdf
• Dasí, À., Pedersen, T., Barakat, L. L., & Alves, T. R. (2021). Teams and project performance: An ability, motivation, and opportunity approach. Project Management Journal, 52(1), 75–89. https://doi.org/10.1177/8756972820953958
• Green, S. G., & Mitchell, T. R. (1979). Attributional processes of leaders in leader-member interactions. Organizational Behavior and Human Performance, 23(3), 429–458. https://doi.org/10.1016/0030-5073(79)90008-4
• Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16(2), 250–279.
• Islam, M. R., Angell, B., Naher, N., et al. (2024). Who is absent and why? Factors affecting doctor absenteeism in Bangladesh. PLOS Global Public Health, 4(4), e0003040. https://doi.org/10.1371/journal.pgph.0003040
• Jiang, K., Lepak, D. P., Hu, J., & Baer, J. C. (2012). How does human resource management influence organizational outcomes? A meta-analytic investigation of mediating mechanisms. Academy of Management Journal, 55(6), 1264–1294. https://doi.org/10.5465/amj.2011.0088
• Joarder, T., Rawal, L. B., Ahmed, S. M., Uddin, A., & Evans, T. G. (2018). Retaining doctors in rural Bangladesh: A policy analysis. International Journal of Health Policy and Management, 7(9), 847–858. https://doi.org/10.15171/ijhpm.2018.37
• Maier, N. R. F. (1955). Psychology in industry: A psychological approach to industrial problems (2nd ed.). Houghton Mifflin.
• McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detecting interactions and moderator effects. Psychological Bulletin, 114(2), 376–390. https://doi.org/10.1037/0033-2909.114.2.376
• Peters, L. H., & O'Connor, E. J. (1980). Situational constraints and work outcomes: The influences of a frequently overlooked construct. Academy of Management Review, 5(3), 391–397. https://doi.org/10.2307/257114
• Purcell, J., & Hutchinson, S. (2007). Front-line managers as agents in the HRM-performance causal chain: Theory, analysis and evidence. Human Resource Management Journal, 17(1), 3–20. https://doi.org/10.1111/j.1748-8583.2007.00022.x
• Purcell, J., Kinnie, N., Hutchinson, S., Rayton, B., & Swart, J. (2003). Understanding the people and performance link: Unlocking the black box. CIPD.
• Sackett, P. R., Gruys, M. L., & Ellingson, J. E. (1998). Ability-personality interactions when predicting job performance. Journal of Applied Psychology, 83(4), 545–556. https://doi.org/10.1037/0021-9010.83.4.545
• Sackett, P. R., Zhang, C., Berry, C. M., & Lievens, F. (2022). Revisiting meta-analytic estimates of validity in personnel selection: Addressing systematic overcorrection for restriction of range. Journal of Applied Psychology, 107(11), 2040–2068. https://doi.org/10.1037/apl0000994
• Siemsen, E., Roth, A. V., & Balasubramanian, S. (2008). How motivation, opportunity, and ability drive knowledge sharing: The constraining-factor model. Journal of Operations Management, 26(3), 426–445. https://doi.org/10.1016/j.jom.2007.09.001
• Vroom, V. H. (1964). Work and motivation. Wiley.

Verification notes: Blumberg and Pringle (1982) was not available in full; its quotations and page numbers are taken from Kellner et al. (2019) and Bos-Nehles et al. (2023), and the lists of its variables from Marin-Garcia and Martinez Tomas (2016). Bailey (1993) was not available; its argument is reported as summarized by later authors. Maier (1955), Vroom (1964) and Campbell et al. (1970) were not consulted directly; the Vroom quotation is as given in Van Iddekinge et al. (2018), and the Campbell et al. point is Kellner et al.'s. Peters and O'Connor (1980), Siemsen et al. (2008), Sackett et al. (1998), Jiang et al. (2012), McClelland and Judd (1993) and Green and Mitchell (1979) are cited for findings stated in their abstracts. The Manufacturing Advantage quotation and study scope are from the book's description on Arne Kalleberg's website. The Bangladesh absence figures come from a single survey conducted in March and April 2002. The comparison table in Section 4, the functional-form table, the diagnostic, both scenarios and the Dhanshiri case are original teaching material for this series.

Next in The Human Capital MBA series: Article 19 examines high-performance work systems and the evidence on whether they pay off.