Human Resource Management

Complete series — 8 parts · Hmmnm!! · hmmnm.in

Human Resource Management Part 1: Scope and the HR Cycle

Aug 20, 2026

In one line: Human Resource Management Part 1 — exam-ready notes in one glance.

In one line: HRM's foundation: Flippo's definition, the managerial-vs-operative split (PODC plus six operative functions), the PM-to-HRM-to-strategic evolution with Ulrich's four roles, and the HR cycle that structures this whole series.

HRM's opening chapter defines the field on four counts. First, the managerial-versus-operational split. Second, the evolution from personnel management to HRM. Third, the functions' cycle. Finally, the modern-era shifts: HR analytics and the gig interface. This note covers the foundation every later part builds on.

Quick Answer: Flippo defines HRM as the planning, organising, directing and controlling of the procurement, development, compensation, integration, maintenance and separation of human resources. Managerial functions run PODC; the six operative functions run procurement to separation. Meanwhile, PM was reactive and cost-minded, whereas HRM is proactive and asset-minded - evolving through the Harvard and Michigan models into strategic HRM (Ulrich's four roles) and today's digital HR. The HR cycle - planning through separation - structures this entire series.

1. HRM: Definition and Nature

• The definition. Flippo: "the planning, organising, directing and controlling of the procurement, development, compensation, integration, maintenance and separation of human resources to the end that individual, organisational and societal objectives are accomplished" - the classic to quote.
• The nature. Five points to list. First, HRM is continuous and pervasive - every manager is an HR manager, hence the line-and-staff duality. Second, it is both an art and a science. Third, it is interdisciplinary: psychology, sociology, economics, law. Finally, it is people-centred and future-oriented.

2. Personnel Management vs HRM vs HRD

• PM vs HRM (the standard table). PM is reactive, treats employees as costs, and bargains through unions with standardised maintenance. In contrast, HRM is proactive, treats people as assets, uses individual contracts, integrates with strategy, and drives commitment - the evolution's core contrast.
• HRD (the subsystem). HRD is the development wing: training, career planning, performance management, and organisational development. Therefore, it is the part of HRM focused on growth and capability - Rao's continuous developmental design. The PM-HRM-HRD three-way is a guaranteed short-answer.

3. The Objectives and the Function Cycle

• The objectives. Four levels: societal (ethically and legally responsive), organisational (the right people for the goals), functional (optimum use), and personal (employees' growth and satisfaction).
• The HR cycle (the operative functions' loop). The loop runs: HR planning; then recruitment; selection; placement and induction; training and development; performance appraisal; compensation; promotion and career; welfare maintenance; finally separation. Moreover, this cycle structures the whole series - each part is one stage. The current-era challenges attach here: retention, engagement, and hybrid work design.

4. The Managerial-vs-Operational Split

• Managerial functions. Planning, organising, directing, controlling - applied to people.
• Operative functions. Six, and every objective question draws on them. First, procurement: planning, recruitment, selection. Second, development: training and careers. Third, compensation: job evaluation, wages, incentives. Fourth, integration: motivation, communication, grievances, IR, discipline. Fifth, maintenance: welfare, health, safety. Finally, separation: retirement, layoff, dismissal.

5. The Modern Evolution

• The eras. The phase ladder runs: scientific management's personnel functions and the WW-era welfare officers; then the 1920s-30s Hawthorne and human-relations turn; next, the 1960s-70s behavioural and systems era; then the 1980s HRM and strategic turn - the Harvard and Michigan models, with Fombrun's 4M match and Harvard's stakeholder map; finally, the 2010s-20s digital-HR and analytics era.
• Strategic HRM. Two fits matter: vertical (strategy-HR alignment) and horizontal (internal consistency). Then Ulrich's model names the four roles: strategic partner, administrative expert, employee champion, change agent - the model to cite.
• The current layer. HR analytics (Part 8's full treatment), the gig and platform workforce (Part 7's statutory interface), employee experience and employer branding, plus AI-recruitment's fairness debates - the modernisation paragraph.

6. How Exams Probe This Topic

• MCQs: Flippo's definition; the six operative functions; PM-versus-HRM contrasts; Ulrich's four roles; HRD's components; the Harvard-Michigan attribution.
• Short answers: HRM's objectives; HRD versus HRM; the HR cycle's stages; the line-and-staff duality.
• Long answers: "Trace the evolution from personnel management to strategic HRM"; "HRM is a pervasive function - discuss with the managerial-operational split."

7. Quick Revision: One-Glance Facts

• Definition. Flippo's procurement-to-separation chain.
• Split. Managerial (PODC) + six operative (procurement, development, compensation, integration, maintenance, separation).
• Evolution. PM, then HRM, then SHRM (Ulrich's four roles), then digital HR.
• HRD. The developmental subsystem: training, careers, OD, performance.

Conclusion. HRM's foundation is one definition (Flippo), one split (managerial versus the six operative functions), and one evolution arc (PM to strategic to digital). Therefore, fix these three, and every later part of this series - recruitment to analytics - slots into a frame the examiner recognises.

Practice Corner: Five Definition Checks (with Answers)

1. Who defined HRM as the planning-through-separation of human resources? - Edwin Flippo.
2. Name the six operative functions. - Procurement, development, compensation, integration, maintenance, separation.
3. Ulrich's four HR roles? - Strategic partner, administrative expert, employee champion, change agent.
4. How does HRD differ from HRM? - HRD is the developmental subsystem: training, careers, OD, performance.
5. What does "pervasive" mean for HRM? - Every manager performs HR functions; it is not only the HR department's work.

The Case Lens: Applying the HR Cycle

Any HRM case - high attrition, a merger integration, a demotivated team - is answered by walking the cycle. First, was procurement right (selection fit, realistic previews)? Then, development (training gaps)? Next, compensation (equity and line of sight)? Then, integration (communication, grievances)? After that, maintenance (welfare, engagement)? Finally, separation (exit interviews feeding back into planning)? Therefore, the cycle is not just a list to recite; it is a diagnostic sequence that locates where the system failed. Meanwhile, examiners consistently reward candidates who use it structurally in long answers - it converts an opinion-laden response into a systems analysis, which is precisely what management papers test.

The Line-and-Staff Distinction (Where Beginners Slip)

HR is a staff function: it advises, designs systems, and services. However, hiring, appraising and directing people remain line responsibilities. The error to avoid - in exams and interviews alike - is treating HR as the department that "does all people management". Instead, the correct formulation: HR designs the machinery (policies, tools, training), while line managers operate it with real employees. This distinction also explains modern frustrations. For example, engagement surveys fail when managers treat them as HR's problem. Therefore, state it explicitly in any HRM answer; it signals professional understanding in the first paragraph.

The HR Function's Case Canon (Three Worked Situations)

The merger-integration case. Two banks merge, and HR inherits duplicate hierarchies, disparate pay scales, and two cultures. Now walk the cycle. Procurement: harmonised selection under a new job architecture. Development: bridge programmes - cross-bank training and leadership alignment. Compensation: grading and benchmarking, within pay-harmonisation's legal and equity constraints. Integration: communication and grievance machinery, anticipating the uncertainty-driven attrition spike. Maintenance: the welfare transition. Separation: the voluntary scheme's design - the VRS's human and financial calculus. The teaching: the cycle is a diagnostic sequence, and each stage's failure looms differently across the merger's lifecycle.

The high-growth startup case. Headcount runs 40 to 400 in two years, and the cycle strains under scale. Procurement: the funnel and the employer brand - professionalising the hiring machinery. Development: manager creation, because the first-line-manager gap is the startup's classic breaking point. Compensation: bands and equity, formalising the ESOP and pay structure. Integration: culture artefacts - codifying the rituals before dilution. Maintenance: the policies' birth, the handbook's first version. Separation: offboarding craft, with a generous alumni network. The lesson: the HR function's maturity must outrun the headcount by six months - the lagging function becomes growth's choking point.

The turnaround case. A loss-making plant puts the cycle under retrenchment's ethics. First, the honest diagnosis: skills redundancy versus demand collapse - different cures. Then, the legal and human discipline of separation, following the Industrial Relations Code's procedures (Part 6's file). Next, survivor-syndrome management - the remaining workforce's guilt and fear. Finally, rebuilding: reskilling and re-engagement. The case completes the cycle's final stage. HRM's hardest work is the goodbye - and the morning after.

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Frequently Asked Questions

What is Flippo's definition of HRM?

The planning, organising, directing and controlling of the procurement, development, compensation, integration, maintenance and separation of human resources - to accomplish individual, organisational and societal objectives.

What are HRM's six operative functions?

Procurement, development, compensation, integration, maintenance, and separation - sitting alongside the managerial functions of planning, organising, directing and controlling.

How do PM and HRM differ?

PM is reactive, cost-minded, union-bargained and maintenance-focused. In contrast, HRM is proactive, asset-minded, individually contracted, strategy-integrated and commitment-driven.

What is HRD's place in HRM?

The developmental subsystem: training, career planning, performance management and organisational development - HRM's growth-and-capability wing.

What are Ulrich's four HR roles?

Strategic partner, administrative expert, employee champion, and change agent - the strategic-HRM model to name in exams.

Why is HRM called pervasive?

Because every line manager performs HR functions daily. HR designs the machinery; line managers operate it - the staff-and-line duality.

Related reading: Human Resource Management Part 3: Training — The ADDIE Route

HRM Part 2: Recruitment and Selection Methods

Aug 20, 2026

In one line: HRM Part 2 — exam-ready notes in one glance.

In one line: Recruitment (the positive pool-building process) feeds selection (the negative rejection process) through a seven-step battery - tests with validity evidence, structured interviews that win on validity, and fairness safeguards now extended to AI screening.

Recruitment and selection questions test four things. First, the recruitment source taxonomy. Second, the selection device battery, with validity evidence. Third, the tests and interview types. Finally, the EEO and fairness layer. This note covers the staffing file.

Quick Answer: Recruitment is the positive process of generating a candidate pool; selection is the negative process of rejection. Sources split internal (promotion, transfer, referrals) and external (advertisements, campuses, agencies, e-recruitment). The seven-step selection sequence runs screening, application blank, tests, interview, references, decision, and medical-offer-induction. Meanwhile, the strongest validity evidence belongs to structured interviews, cognitive tests and work samples. Finally, the modern layer adds AI-screening fairness and DPDP consent.

1. Recruitment vs Selection

• Recruitment. The positive process: searching for and stimulating candidates to generate a pool. In contrast, selection is the negative process of rejection - choosing from the pool. Recruitment precedes and feeds selection: the definition pair every paper opens with.
• The process context. Staffing anchors in HR planning - demand-supply forecasting finds the gap, and the gap drives the recruitment plan. In addition, job analysis supplies the job description and specification - the documents every staffing decision references.

2. The Sources: Internal and External

• Internal sources. Promotion, transfer, job posting, employee referrals, and former employees' rehire. Merits: motivation, low cost, and known performance. However, the demerits matter too: inbreeding, jealousy, and the vacancy chain each promotion creates.
• External sources. Advertisements, employment exchanges, campus recruitment, private agencies and headhunters, walk-ins, e-recruitment (the job-portal and LinkedIn era), labour contractors, poaching and raiding, plus webinars and hackathons. Merits: fresh blood and wider choice. Demerits: cost, adjustment time, and demotivated insiders.
• The evaluation metrics. The modern answer adds yield ratios, cost-per-hire, time-to-fill, and quality of hire.

3. The Selection Process

The seven-step standard (the numbered answer):

1. Preliminary screening - the application review.
2. Application blank - including the weighted-application variant, where biodata gets predictive weights.
3. Selection tests (section 4).
4. Employment interview (section 5).
5. Reference and background checks - with their validity caveats.
6. Selection decision - the offer and negotiation.
7. Physical or medical examination, job offer, placement and induction. Furthermore, induction and orientation act as the bridging programme - socialisation and the realistic job preview reduce early attrition.

4. The Tests Battery

• The test types. First, aptitude (potential: verbal, numerical, reasoning). Then achievement (learned skills) and personality (Big-Five-type inventories; projective tests like the TAT and Rorschach). Next, interest tests (the Strong and Kuder type) and integrity tests (honesty measures as polygraph alternatives). Finally, motor and physical-ability tests, and assessment centres with multiple simulations - the type-definition matches.
• The validity criteria (a quality test-set). First, reliability - consistency via test-retest and split-half. Then validity: predictive or criterion, content, and construct. In addition, standardisation and norms, fairness, and usability - the five-criteria MCQ.
• The evidence hierarchy. Structured interviews, cognitive-ability tests, and work samples show the strongest criterion validity. However, graphology stands debunked - the evidence-based line for depth.

5. The Interview and Its Errors

• The types. The structured interview - fixed questions and scoring - is the validity winner. Then unstructured and semi-structured forms. In addition: panel interviews, stress interviews (deliberate pressure, with the ethics debate attached), behavioural interviews (STAR: Situation-Task-Action-Result), and situational interviews ("what would you do?").
• The error catalogue (the appraisal-bias crossover). Halo, horns, first-impression or primacy, contrast, similar-to-me, stereotyping, non-verbal bias, and anchoring on irrelevant cues. The fixes mirror Part 4's list: structure, multiple interviewers, behaviour-anchored scoring, and training.
• The modern layer. AI-video interview platforms raise the algorithmic-fairness and DPDP-consent debates. Meanwhile, gamified assessments and asynchronous interview formats spread - the current paragraph.

6. How Exams Probe This Topic

• MCQs: the recruitment-versus-selection polarity; source classifications; test-validity types; STAR's letters; interview-error definitions; the seven-step order.
• Short answers: internal-versus-external merits; the selection-test criteria; structured interviews' superiority.
• Cases: design a selection process for a role, with the method choice justified; diagnose a biased interview process.

7. Quick Revision: One-Glance Facts

• Pair. Recruitment = positive (pool); selection = negative (rejection).
• Steps. Screening, blank, tests, interview, references, decision, medical, induction.
• Tests. Aptitude, achievement, personality, interest, integrity; validity: predictive, content, construct.
• Interview. Structured wins; STAR; the halo-contrast-similar-to-me errors.

Conclusion. Staffing answers run on the positive-negative pair, the seven-step sequence, the test battery's validity criteria, and the interview error catalogue. Therefore, anchor method choices in validity evidence, and mirror the appraisal-bias fixes onto interviews. With that, the second HRM chapter is fully covered.

Practice Corner: Five Selection Checks (with Answers)

1. What are the three test-validity types? - Predictive (criterion), content and construct.
2. Which interview type shows the highest validity? - The structured interview.
3. What does STAR stand for? - Situation, Task, Action, Result.
4. Which error is "rating everyone average"? - Central tendency.
5. What is the "you can eliminate the middleman but not his work" idea? - Removing intermediaries does not remove the channel functions.

The Case Method: Selection Process Design for a Sales Role

Given: a pharma company hiring 50 medical representatives, with high first-year attrition. The strong answer walks the seven steps with justifications. First, screen on application blanks with weighted biodata - past sales performance predicts. Then run a cognitive-plus-personality battery, chosen for validity evidence over unstructured judgment. Next, a structured behavioural interview with STAR scoring - the validity winner. In addition, a realistic job preview including travel demands, cutting attrition through self-selection. Then reference checks on finalists only. Finally, a three-month onboarding with early-attrition tracking as the process's own metric. However, the weak answer lists tools without the validity logic and without the attrition-specific design. Meanwhile, every staffing case rewards the same pairing: the process sequence plus the reason each element earns its place.

The Realistic-Job-Preview Case (The Attrition Lever)

A BPO hires 200 agents; 40% quit within 90 days. The diagnosis candidates should reach: the selection process sold a rose-tinted job. Consequently, the fix runs inside selection itself. First, a realistic job preview - shift timings, a call-pressure simulation, a "worst day at work" exercise - lets mismatched candidates self-select out. Then a work-sample test (a live mock call) replaces generic interviews. Finally, post-offer engagement - a buddy and a preview visit - cuts the offer-to-joining drop. The redesign costs a fraction of one month's attrition. Therefore, the principle for any staffing case: it is cheaper to lose a wrong candidate at selection than a right one at day thirty.

The Digital-Recruitment Dilemmas (AI Meets Hiring)

Three cases define the ethics-and-design intersection. First, resume-screening bias: Amazon's 2018 tool, trained on male-dominated history, learned to penalise women's resumes - and was scrapped. The lesson: biased input reproduces biased output. Second, video-interview analytics: facial and speech scoring raises validity questions, plus DPDP consent and purpose-limitation duties - candidates hold the right to explanation. Third, the design principles: bias audits testing whether flag rates differ by protected group, human review for adverse decisions, and transparency notice. Therefore, "AI in recruitment" answers - new NET and MBA favourites - run these cases plus these principles.

The Validity Evidence (The Numbers That Persuade)

The evidence-based answer's backbone is a small validity table (correlations with job performance). First, the structured interview: about 0.5, against roughly 0.2 unstructured - structure yields a 2.5-times improvement. Then cognitive ability: about 0.5 - with the complement-versus-substitute debate resolved by additive evidence. Next, work samples: 0.5-plus - the best but costliest, fitting high-stakes roles. Meanwhile, integrity tests reach about 0.4, and personality's conscientiousness carries roughly 0.3 - the Big Five's one validity carrier. Finally, graphology: about zero - the debunked cautionary tale. These five numbers give "evaluate selection methods" answers their scientific spine - and the evidence-based-HR claim its substance.

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Frequently Asked Questions

What is the difference between recruitment and selection?

Recruitment is the positive process of building a candidate pool. In contrast, selection is the negative process of rejecting from that pool. Recruitment precedes and feeds selection.

What are the seven selection steps?

Preliminary screening, application blank, selection tests, employment interview, reference checks, selection decision, then medical examination, offer, placement and induction.

Which selection methods have the strongest validity?

Structured interviews, cognitive-ability tests, and work samples - each around 0.5 correlation with job performance. Graphology sits near zero, thoroughly debunked.

What does STAR stand for in behavioural interviews?

Situation, Task, Action, Result - the structured storytelling frame that makes behavioural interviewing scoreable.

What are the test-validity types?

Predictive or criterion validity, content validity, and construct validity - backed by reliability (test-retest, split-half), standardisation, fairness and usability.

How does AI recruitment raise fairness issues?

Trained on biased history, screening tools reproduce it - Amazon's scrapped 2018 tool is the canonical case. Therefore: bias audits, human review of adverse decisions, and DPDP consent with explanation rights.

Related reading: Human Resource Management Part 3: Training — The ADDIE Route · HRM Part 4: Performance Appraisal Methods and Biases · HRM Part 5: Compensation Structure and Incentives

Human Resource Management Part 3: Training — The ADDIE Route

Aug 20, 2026

In one line: Human Resource Management Part 3 — exam-ready notes in one glance.

In one line: Training and development runs on ADDIE's five stages, the on-job-versus-off-job method mix matched to skill type, and Kirkpatrick's four evaluation levels plus ROI - with 70-20-10 as the modern heuristic.

Training and development questions test three things. First, the training process models, ADDIE above all. Second, the methods - on-job versus off-job, with named techniques. Third, the evaluation levels, Kirkpatrick's four. This note covers the full T&D file.

Quick Answer: ADDIE is the process model: Analyze (the three-level TNA), Design, Develop, Implement, Evaluate. Methods split on-job (JIT, coaching, mentoring, rotation) and off-job (cases, role play, simulations, in-basket, vestibule, T-groups). Evaluation climbs Kirkpatrick's ladder - Reaction, Learning, Behaviour, Results - extended by ROI. Moreover, Knowles' andragogy grounds adult learning design, and the 70-20-10 model frames the continuous-learning culture.

1. Training vs Development vs Education

• Training. Present-job-linked, skill-focused, narrow and specific - operational.
• Development. Future-oriented, growing the whole person - managerial and career-linked. This is the standard pair.
• Education. Broad theoretical preparation through institutions. Together, the three-way distinction opens most answers.
• The systemic context. Training sits inside HRD - the integrated frame with performance management, career planning and organisational development, plus the HRD-audit and climate concepts.

2. The Training Process and ADDIE

• ADDIE (the model to reproduce). First, Analyze: training-needs analysis at three levels - organisational, task, person. Second, Design: objectives in measurable terms, learner profile, method-mix. Third, Develop: materials, modules, trainers. Fourth, Implement: delivery. Finally, Evaluate - Kirkpatrick, covered in section 4.
• The classic Indian-frame companion. The systematic training cycle: identify needs, set objectives, design, deliver, evaluate, then feed back into needs. In addition, list the training-need spotters: performance gaps, new technology, expansion, and problem areas.
• The design choices. Objectives run SMART. Then note trainee readiness and motivation: Knowles' andragogy assumes self-direction, experience-richness, and problem-centredness - the theory name-drop. Finally, transfer-of-learning design uses identical elements and general principles.

3. Methods: On-Job and Off-Job

• On the job (learn while doing). First, job instruction training - JIT's four steps: prepare, instruct, try out, follow up. Then coaching (one-on-one guidance) and mentoring (broader: career and wisdom). Next, job rotation for cross-experience, and understudy or assistantship. Finally, apprenticeship (the NAPS-type statutory layer) and committee assignments.
• Off the job. Lectures and seminars come first. Then the case study - the decision-practice method and MBA staple. Next, role play for interpersonal practice, management games and simulations for competitive decision environments, and in-basket exercises for priority drills. In addition: sensitivity or T-group training (interpersonal awareness, with the OD connection), vestibule training (off-line simulated workstations - machinery skills without production risk), e-learning and MOOCs as the modern base, and behavioural modelling (demonstration-practice-feedback for social skills).
• The matching logic case answers need. Motor skills pair with vestibule or JIT. Interpersonal skills pair with role play or modelling. Meanwhile, decision skills pair with cases or in-basket. Finally, conceptual skills pair with lectures and simulations.

4. Evaluating Training: Kirkpatrick and Beyond

• The four levels - HRM's most tested numbered list. (1) Reaction: did trainees like it - post-programme sheets. (2) Learning: did they learn - pre-post tests. (3) Behaviour: did on-job behaviour change - 360-degree follow-ups at 3-6 months. (4) Results: organisational outcomes - productivity, quality, retention. Then the cost-benefit extension: ROI = (net benefit − cost) / cost × 100 - the numerical they ask.
• The frameworks beyond. Hamblin's five levels add ultimate value; CIRO runs context-input-reaction-outcome; Phillips adds the ROI fifth level - name-drops for depth.
• The honest critique line. Most organisations stop at level 1 or 2. However, behaviour-results measurement is rare - and that gap makes evaluation questions essay-worthy.

5. The Modern Layer

• Digital-first learning: LMS platforms, micro-learning (short-burst mobile content), and AI-personalised paths.
• The reskilling imperative: automation- and AI-driven TNA at enterprise scale - the corporate academies. Meanwhile, the continuous-learning culture rests on the 70-20-10 model: 70% experience, 20% coaching and exposure, 10% formal training - the heuristic every paper now cites.
• The Indian statutory hooks. The Apprentice Act's amendments, plus the Skill India and PMKVY ecosystem - the national skilling tie-in for NET-type papers.

6. How Exams Probe This Topic

• MCQs: ADDIE's order; Kirkpatrick's four levels and which level a described measure belongs to; method matches (vestibule-machinery, in-basket-prioritisation); 70-20-10; andragogy's assumptions; TNA's three levels.
• Short answers: training versus development; on- versus off-job methods; transfer-of-learning conditions.
• Cases: design a training programme (ADDIE walked through a described gap); an evaluation plan across the four levels.

7. Quick Revision: One-Glance Facts

• Process. TNA (organisation-task-person), then ADDIE's five stages.
• Methods. On-job: JIT, coaching, mentoring, rotation. Off-job: case, role play, simulation, in-basket, vestibule, T-group. Plus the skill-method match.
• Evaluation. Kirkpatrick 1-4 plus ROI; Hamblin and Phillips extensions; the stop-at-reaction critique.
• Modern. 70-20-10; micro-learning; reskilling; PMKVY.

Conclusion. T&D answers are model answers: ADDIE for the process, the method-mix matched to skill type, and Kirkpatrick's ladder plus ROI for evaluation. Therefore, fix the three numbered lists - ADDIE-5, Kirkpatrick-4, TNA-3 - and the 70-20-10 heuristic. With those, every training question assembles from this note directly.

Practice Corner: Five Method Checks (with Answers)

1. ADDIE's five stages? - Analyze, Design, Develop, Implement, Evaluate.
2. TNA's three levels? - Organisational, task, person.
3. Kirkpatrick's four levels? - Reaction, Learning, Behaviour, Results.
4. The adult-learning theory's author? - Knowles (andragogy).
5. The experience-learning heuristic? - 70-20-10 (experience, exposure, formal training).

The Training Case Frame (Any "Design a Programme" Question)

Take a described gap - say, bank officers failing to sell new digital products - and walk six moves. First, TNA: is the gap knowledge, skill, or motivation? Each has a different remedy. Second, objectives in measurable terms: "handle 90% of customer queries without escalation within 30 days." Third, method-matched design: product knowledge through e-learning modules; objection handling through role play; system fluency through vestibule simulation. Fourth, delivery on andragogy principles - problem-centred and experience-honouring. Fifth, Kirkpatrick-planned evaluation: tests at level 2, supervisor observation at level 3, sales figures at level 4. Finally, the transfer-support layer - job aids, coaching, incentive alignment. That is the step most answers omit, and most real programmes fail on. Therefore, six moves with named frameworks attached is a full-mark answer in any HRM paper.

The Transfer-of-Training Layer (Where Training Actually Fails)

Most training fails after the classroom - the transfer problem. Meanwhile, the exam-aware answer names its mechanics. The trainee returns to a workplace where the supervisor never heard of the new method. Furthermore, the incentive system still rewards the old one, and the workload forgives no practice curve. The research-backed countermeasures, by phase: before training, the supervisor's goal-setting session and the trainee's involvement in needs analysis build commitment early. During, behaviour modelling and practice-with-feedback beat lecture - the learning-by-doing margin. After, relapse prevention identifies the triggers pulling the trainee back to old habits; action plans carry follow-up dates; and the supervisor's reinforcement covers the first ninety days. Therefore, every training case answer must extend beyond programme design to the transfer climate. The question "will this training change on-job behaviour?" is answered by the organisation's design, not the trainer's.

The Training ROI Calculation (The Numerical That Appears)

A programme costs Rs 8,00,000 - design Rs 2 lakh, delivery Rs 4 lakh, trainee time at cost Rs 2 lakh - for 40 employees whose average error rate costs Rs 25,000 each annually. Suppose the training halves error losses for one year. Then savings equal 40 × 12,500 = Rs 5,00,000, and first-year ROI = (5L − 8L)/8L = −37.5%: the programme looks failed at level 4. However, if the effect persists two years - and skills do - cumulative savings reach Rs 10 lakh, and ROI turns +25%. In addition, add the retention effect: four avoided resignations × Rs 1.5 lakh replacement cost = Rs 6 lakh, and the level-4 case clears 100%. The examinable lessons: first, ROI's denominator must carry trainee time - the hidden cost candidates forget. Second, benefits must be time-boxed - one year versus skill-life. Finally, the level-3 measurement design decides whether benefits are countable at all. Consequently, most firms cannot prove the savings they believe they got.

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Frequently Asked Questions

What is ADDIE?

The training-process model: Analyze (three-level TNA), Design (measurable objectives, method mix), Develop (materials and trainers), Implement (delivery), Evaluate (Kirkpatrick).

What are Kirkpatrick's four levels?

Reaction (did they like it), Learning (pre-post tests), Behaviour (on-job change at 3-6 months), Results (organisational outcomes). Phillips adds ROI as the fifth level.

What is the difference between training and development?

Training is present-job-linked, skill-focused and narrow. In contrast, development is future-oriented growth of the whole person - managerial and career-linked.

What is andragogy?

Knowles' adult-learning theory: adults are self-directed, experience-rich, and problem-centred. These assumptions shape training design.

What is vestibule training?

Off-job training on simulated workstations away from the production line - machinery skills learned without production risk.

What is the 70-20-10 model?

The learning heuristic: 70% from on-job experience, 20% from coaching and exposure, 10% from formal training - the continuous-learning-culture frame.

HRM Part 4: Performance Appraisal Methods and Biases

Aug 20, 2026

In one line: Performance appraisal tests the traditional-versus-modern methods (MBO, 360-degree and BARS as centrepieces) and the rater-bias catalogue - halo, central tendency and friends - each with its evidence-backed fix.

Performance appraisal questions test two things. First, the traditional versus modern methods, with MBO, 360-degree and BARS as the centrepieces. Second, the rater-bias catalogue - halo, central tendency and friends. This note covers the methods and the error set with exam precision.

Quick Answer: Appraisal serves administrative purposes (pay, promotion, retention) and developmental ones (strengths, gaps, career planning). Traditional methods run from ranking to the graphic rating scale; modern ones are MBO (Drucker, 1954), 360-degree appraisal, BARS, and the assessment centre. Meanwhile, the bias catalogue covers halo, central tendency, leniency, recency, contrast, similar-to-me and stereotyping. The fixes: frame-of-reference training, BARS, multiple raters, incident logs, and calibration meetings. Finally, the modern reframe turns annual appraisal into continuous performance management.

1. Appraisal Purposes and the Process

• The purposes. Two families open most answers. First, administrative: salary, promotion, transfer, and retention decisions. Second, developmental: strengths and gaps, training needs, and career planning. In addition, appraisal validates selection and feeds HR planning.
• The process. The sequence runs: establish standards; communicate them; measure with the chosen method; compare; discuss in the interview; decide and act; then feed back into standards. However, note the modern reframe: appraisal becomes performance management - a continuous goal-setting, coaching and review cycle, not an annual event. That distinction is a favourite exam question.

2. Traditional Methods

1. Ranking - simple ordinal comparison; however, it fails at absolute levels.
2. Paired comparison - every employee against every other: n(n-1)/2 judgments.
3. Grading - predefined categories like A-B-C.
4. Forced distribution - fitting ratings to a bell (the GE vitality-curve lineage, with the fairness critique attached).
5. Checklist - yes/no behaviour items, with weighted variants.
6. Critical incident method (Flanagan) - specific good and bad behavioural events, recorded contemporaneously. Rich, but alone it offers no quantification.
7. Graphic rating scale - traits rated on a continuum. It is the most used traditional form - and the most bias-prone.

3. Modern Methods: MBO, 360-Degree, BARS

• MBO (Drucker, 1954 - "The Practice of Management"). The joint goal-setting cascade runs: organisational, then departmental, then individual SMART goals, with periodic review and participative evaluation. Strengths: objectivity and commitment. However, the weaknesses earn marks too: goal short-termism, paperwork, and shaky assumptions about measurability and participation - the balanced critique.
• 360-degree appraisal. Ratings come from supervisors, peers, subordinates, self, and customers or internal clients - the multi-source logic. Strengths: developmental richness and bias dilution. Weaknesses: mutual back-scratching, trust and anonymity requirements, and unsuitability for administrative pay decisions. Therefore, the consensus line: best as developmental input.
• BARS (Behaviourally Anchored Rating Scales). Scales are anchored with specific behavioural examples, built from critical incidents plus scaling. Consequently, BARS combines the rating scale's quantification with the critical incident's job-relatedness. The construction steps - collect incidents, cluster into dimensions, scale-anchor - form a short-answer staple. However, the cost is development time.
• The assessment centre method - multiple simulations and multiple assessors, measuring potential for managerial succession. In addition, forced-choice and essay methods complete the list.

4. The Rater-Bias Catalogue

The error set (match-the-definition MCQs appear every year):

1. Halo error - one positive trait colours all ratings; its opposite is horns.
2. Central tendency - everyone rated average; commitment avoided.
3. Leniency / severity - systematic over- or under-rating.
4. Recency effect - the last weeks dominate the year.
5. Primacy effect - first impressions anchor.
6. Contrast error - rated against the previous candidate, not the standard.
7. Similar-to-me bias - favouring resembling ratees.
8. Stereotyping - group-based attribution.

• The counter-measures. First, rater training - frame-of-reference training is the evidence-backed fix. Then behaviour-based instruments (BARS), multiple raters (360-degree), diary and critical-incident logs against recency, and calibration meetings - the fixes list for long answers.

5. The Appraisal-Feedback Interview

• The three approaches. Tell-and-sell (persuasive), tell-and-listen (problem-solving), and problem-solving (developmental) - the modern preference.
• The conduct principles. Six points: specific behaviours, not personalities; no surprises (feedback through the year); employee participation; forward-looking development plans (the IDP - individual development plan); documented follow-up; and consistency.
• The legal-fairness layer. Documented, job-related, consistent criteria - because appraisal becomes evidence in disputes.

6. How Exams Probe This Topic

• MCQs: method matches (MBO-Drucker; critical incident-Flanagan); BARS anchors; bias definitions; 360-degree's sources; forced-distribution mechanics; assessment-centre features.
• Short answers: traditional versus modern methods; MBO's process and weaknesses; bias remedies.
• Cases: diagnose a described appraisal failure (a bias named and fixed); design an appraisal system for a described firm, with the method choice justified.

7. Quick Revision: One-Glance Facts

• Methods. Traditional seven (ranking to graphic scale); modern four (MBO, 360-degree, BARS, assessment centre).
• MBO. Drucker 1954; SMART cascade; participation.
• BARS. Behavioural anchors from critical incidents.
• Biases. Halo, central tendency, leniency-severity, recency-primacy, contrast, similar-to-me, stereotyping - fixed by FOR training, BARS, multi-raters, logs, calibration.
• Interview. Tell-and-sell / tell-and-listen / problem-solving.

Conclusion. Appraisal answers run on two lists - the methods ladder (traditional to MBO-360-BARS) and the bias catalogue - plus the modern reframe into continuous performance management. Therefore, match each bias to its fix, name Drucker and Flanagan correctly, and this HRM staple is fully banked.

Practice Corner: Five Bias Checks (with Answers)

1. "One good trait colours all ratings" - which error? - Halo.
2. "Everyone gets a 3/5" - Central tendency.
3. Which method combines quantification with behavioural anchors? - BARS.
4. Who created MBO? - Drucker (1954).
5. The best developmental instrument using multiple raters? - 360-degree appraisal.

The Bias-Fix Table (Memorise as Pairs)

Halo/horns pairs with behaviour-anchored scales and multiple dimensions. Central tendency pairs with forced distribution or calibration meetings. Leniency and severity pair with frame-of-reference training using benchmarks. Recency pairs with the critical-incident log maintained through the year. Contrast pairs with rating against standards, not the previous candidate. Similarly, similar-to-me pairs with structured criteria and diverse rating panels. Every appraisal case presents two or three of these errors in a vignette. Therefore, the complete answer names each error, cites its fix from this table, and closes with the institutional layer - FOR training, BARS, and the calibration cycle - that prevents recurrence. Six pairs, one table: the entire error side of the syllabus in retrievable form.

The Modern Methods Extended (How Each Actually Works)

MBO's full mechanics go beyond the label. First, the goal-setting meeting: the subordinate proposes, the manager refines - participation is the motivation engine, not a formality. Then the periodic review's course correction, not the annual ambush. Finally, appraisal against agreed - not assumed - criteria. However, note the documented failure modes: goals set easy (sandbagging), goals set for the measurable alone (what gets counted gets done - and nothing else does), and reward linkage that converts exploration into target-chasing. Meanwhile, the sophisticated answer praises MBO's philosophy while noting its practice decayed into appraisal-by-objective. That is exactly why the OKR revival explicitly separates aspirational objectives from performance-rated compensation.

360-degree's implementation details decide its fate. First, rater selection: self-chosen raters inflate, while system-assigned raters alienate. Second, anonymity thresholds: below three raters per category, anonymity collapses - and candour dies with it. Third, feedback facilitation: raw data handed to an unprepared recipient wounds, whereas facilitated debriefs develop. Finally, the firewall rule: development-only 360s stay out of pay decisions, because the moment they enter, every rater becomes a politician. Meanwhile, the Indian corporate record: adopted widely since the 2000s IT-sector wave, genuinely functioning in perhaps one implementation in three - a gap candidates can demonstrate with the firewall rule alone.

BARS' construction, honestly costed: critical incidents collected from subject-matter experts by the hundreds, clustered into dimensions, then anchored at scale points by behavioural consensus. Consequently, each job family takes six to twelve months to develop - which is why BARS lives in textbooks more than appraisal systems. Its exam uses: the construction steps as a short answer, and the point-method-versus-BARS trade (generalisability versus job-relatedness) as the comparison question.

The Appraisal-Interview Script (The Manager's Side)

The interview that develops rather than defends has five moves. First, open with the employee's self-assessment - the manager's information doubles while defensiveness halves. Second, discuss behaviours against agreed standards, using diary and incident evidence, never adjectives. Third, co-author the development plan: the IDP with two commitments, one capability and one visibility. Then close with the follow-up date - the review that makes the plan real. Finally, document everything. Five moves, twenty minutes - and the same script answers any case phrased as "how should the manager conduct the appraisal discussion?"

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Frequently Asked Questions

What are the modern appraisal methods?

Four: MBO (Drucker, 1954 - SMART goal cascades with participation), 360-degree appraisal (multi-source ratings), BARS (behaviourally anchored scales), and the assessment centre (simulations measuring potential).

What is halo error?

One positive trait colours all ratings of a person; its negative twin is horns error. The fix: behaviour-anchored scales rating multiple dimensions separately.

What is central tendency error?

Rating everyone average to avoid commitment. Therefore, the fixes are forced distribution or calibration meetings that force rating discipline.

How is BARS constructed?

Collect critical incidents from subject-matter experts, cluster them into dimensions, then anchor scale points with behavioural consensus. Costly - six to twelve months per job family - but job-related.

Why should 360-degree appraisal stay out of pay decisions?

Because once ratings affect pay, raters become political - the firewall rule. Development-only 360s preserve anonymity, candour, and developmental value.

What is frame-of-reference training?

Rater training on shared performance standards with benchmark examples - the evidence-backed fix for leniency, severity, and inconsistent rating.

Related reading: HRM Part 2: Recruitment and Selection Methods · HRM Part 5: Compensation Structure and Incentives

HRM Part 5: Compensation Structure and Incentives

Aug 20, 2026

In one line: HRM Part 5 — exam-ready notes in one glance.

In one line: Compensation management runs on the total-rewards frame (base + incentives + benefits + LTIs), internal-and-external equity structures, the Code on Wages 2019's statutory floor, and ESOP alignment — anchored in Adams' equity and Vroom's expectancy theories.

Compensation questions test four layers. First, the base salary structures: grades and pay scales. Second, the statutory wage machinery: minimum wages and the Code on Wages 2019. Third, the incentive families, individual and group. Finally, the executive-ESOP layer. This note assembles the complete compensation file, with worked cases and one-glance revision facts at the end.

Quick Answer: Total rewards combine base pay, dearness-type adjustments, performance incentives, benefits and perquisites, and long-term incentives. The base rests on two equities: job evaluation internally, pay surveys externally. Meanwhile, the Code on Wages 2019 subsumes four laws (Payment of Wages 1936, Minimum Wages 1948, Payment of Bonus 1965, Equal Remuneration 1976), sets a floor wage, and fixes a uniform wage definition. Incentives span piece rates, commissions, gainsharing (Scanlon-Rucker-Improshare) and ESOPs. Finally, Adams' equity and Vroom's expectancy theories anchor every design answer.

1. Compensation: The Components and Objectives

Compensation is not a single number on a payslip; it is an architecture. Understanding the architecture first makes every later topic — statutes, incentives, ESOPs — easier to place.

• The components. The total-rewards frame stacks five parts: base pay (the fixed, contracted salary); dearness-allowance-type adjustments (inflation-linked corrections, historically important in Indian pay); incentives (performance-linked variable pay); benefits and perquisites (statutory and voluntary — the cafeteria approach lets employees choose from a menu); and long-term incentives (ESOPs, RSUs, performance shares).
• The objectives. Five, for the opening list: attract, retain, and motivate; internal and external equity; legal compliance; cost-effectiveness; and strategic alignment of pay with performance. Note that these objectives frequently tension against each other — cost-effectiveness pulls against attraction, and internal equity can pull against market-rate hiring. Naming the tension is what separates a good answer from a great one.
• The equity theories backing it. First, Adams' equity theory: employees compare their input-output ratios with referents (peers, market counterparts), and perceived under-reward creates tension that demands correction — through demands, reduced effort, or exit. Then, Vroom's expectancy: motivation = expectancy × instrumentality × valence; effort, performance, reward, valence — incentive credibility requires the instrumentality link to be visible and trusted. Finally, agency theory frames executive pay: principals (shareholders) use pay contracts to align agents (managers) whose interests would otherwise diverge — the anchors for long answers.

2. Base-Pay Structures

• Job evaluation determines systematic relative worth of jobs within the organisation. The methods ladder runs: ranking, classification/grading, factor comparison, and the point method. Ranking is quick but crude; classification slots jobs into predefined grades; factor comparison benchmarks jobs factor-by-factor against key jobs; and the point method — compensable factors, times degrees, times weights — is the most defensible and therefore the internal-equity instrument of choice.
• Pay surveys provide the external-equity instrument through benchmarking against market comparators. Then the pay line and grades (range spreads, midpoints, overlaps) get built from the two equities together: job evaluation orders jobs internally; surveys price them externally; the grade structure reconciles both. A typical range spread runs 30–50% from minimum to maximum, with successive ranges overlapping to allow growth without promotion.
• The structural vocabulary. Broadbanding means fewer, wider bands — the flexibility trend that de-emphasises titles and supports lateral movement. Meanwhile, pay compression is the new-hire-versus-veteran inversion (worked through in the case below). In addition, skill-and-competency-based pay rewards capability acquired, not the job held — it fits flatter, project-based organisations.
• The tenure frames. Seniority-based pay is predictable, low-conflict and administratively simple, but demotivating to performers. In contrast, performance-based pay motivates but stays contestable — ratings get disputed, and measurement noise can punish good work. The classic contrast: predictability versus motivation.

3. The Statutory Layer: The Code on Wages

• The Code on Wages 2019 is enacted, with implementation pending final rules and notifications — always check current status before quoting. It subsumes four laws: the Payment of Wages Act 1936, Minimum Wages Act 1948, Payment of Bonus Act 1965, and Equal Remuneration Act 1976 — the subsumption MCQ that appears most reliably.
• The key provisions. First, universal coverage removes the earlier wage-ceiling filter, extending protection to all employees. Then comes the uniform definition of wages — allowances cannot indefinitely be excluded, effectively making basic-plus-dearness roughly 50% of total remuneration; this directly inflates PF and gratuity computations for allowance-heavy pay structures. Next, a statutory floor wage set by the Centre (states cannot fix minimums below it), timely payment obligations, and prohibitions on arbitrary deductions. Finally, gender-neutral equal remuneration — notably broadened beyond "male and female" — and liberalised bonus ceilings.
• The adjacents. The Payment of Gratuity Act 1972 (five years' continuous service for eligibility; formula: 15/26 × last drawn wages × completed years of service, capped by the notified ceiling). Then the EPF & MP Act 1952 (12% employer-employee contribution structure on statutory wages). Finally, the Code on Social Security 2020 — the statutory family for answer depth, extending coverage toward gig and platform workers.

4. Incentive Systems

• Individual incentives. First, piece rates — including Taylor's differential piece rate, which pays a higher rate above standard and a lower rate below — the scientific-management root of all incentive design. Then time-saving bonuses (Halsey, Rowan-style premium plans), commission (sales' standard), and merit pay and increments merged into base salary, plus performance bonuses as the non-accumulating variable-pay layer. The distinction matters: merit pay compounds into base; bonuses reset each cycle.
• Group and organisation-wide incentives. Gainsharing leads — the Scanlon (ratio of payroll cost to sales value), Rucker (payroll to production value) and Improshare (standard-hours versus actual-hours) productivity-sharing plans. Then profit sharing (organisation-level, deferred or current). Finally, employee stock ownership — the group ladder's long-horizon rung.
• The design principles. Line of sight (the effort-to-reward link must be visible), controllable performance measurement (never pay on what the employee cannot influence), frequency (payout close to the performance), and cap-and-floor structures to bound extremes. However, hold the cautions: gaming of metrics, quality erosion when quantity alone is paid, and the crowding-out of intrinsic motivation — the behavioural-economics critique for depth answers.
• The indirect layer. Fringe benefits and flexible cafeteria plans (tax-efficient benefit choice); statutory benefits (PF, gratuity, ESI); and work-life benefits (flexibility, leave, wellness) complete total rewards.

5. Executive Pay, ESOPs and the New-Age Layer

• Executive compensation. The stack: base + annual bonus + long-term incentives (ESOPs, RSUs, performance shares) + perquisites + severance (golden parachutes). Meanwhile, the say-on-pay and disclosure debates — including SEBI's remuneration-disclosure norms for listed boards — give the governance cross-link. The agency-theory question always lurks: does the contract align the agent, or merely enrich them?
• ESOPs. The mechanics run grant → vesting → exercise, with cliff vesting (nothing until a threshold date) and graded vesting (portions vest over time). The rationale is retention and shareholder alignment. Furthermore, fair-value expensing governs the accounting, while dilution and repricing draw the standard criticism. Indeed, the startup-ESOP culture — liquidity programmes and buybacks letting early employees cash out — is the modern Indian case layer.
• The gig-era layer. The Code on Social Security 2020's platform-worker frame, and the unresolved challenge of benefits for the non-employee workforce — the frontier paragraph for case answers.

6. How Exams Probe This Topic

• MCQs: the four subsumed acts; the gratuity formula and eligibility; job-evaluation methods; the Scanlon/gainsharing family; equity-expectancy attribution; broadbanding; ESOP mechanics; the uniform wage definition.
• Short answers: compensation objectives; individual versus group incentives; merits of the point method; ESOP pros and cons.
• Cases: design a compensation plan for a described firm, justified with equity theory; fix a demotivating incentive via line-of-sight diagnosis; the Code-on-Wages impact question.

7. Quick Revision: One-Glance Facts

• Frame. Total rewards = base + incentives + benefits + LTIs; objectives: attract-retain-motivate + equity + compliance + cost-effectiveness + strategic alignment.
• Structures. Job evaluation (point method as gold standard) + surveys → pay line → grades; broadbanding; skill-based pay; pay compression.
• Statute. Code on Wages 2019 = four acts subsumed; floor wage; uniform definition (~50% rule); gratuity = 15/26 × last drawn × years, after 5 years.
• Incentives. Piece-commission-merit (individual); Scanlon-Rucker-Improshare and profit share (group); ESOPs (grant → vest → exercise); design on line of sight.
• Theory. Adams' equity, Vroom's expectancy, agency theory.

Conclusion. Compensation is equity-engineered motivation: internal fairness through job evaluation, external fairness through surveys, line-of-sight incentives for the variable layer, the Code on Wages for the statutory floor, and ESOPs for alignment. Therefore, anchor every design answer in equity-expectancy theory, and hold the gratuity formula and the four subsumed acts cold — that is the complete compensation answer.

The Compensation Design Case (End-to-End Worked)

A 300-person product startup gets poached by a rival's 30% offers. First, diagnose: is it pay level or pay structure? The survey says the level matches market median. However, the rival pays purely on performance, while this firm's 80-20 fixed-variable split cushions the average. Then redesign. Fix base at the 50th percentile, but shift variable pay to 25% at target with a 1.5x cap for outperformance. Consequently, top performers' expected value beats the rival, while the average performer's smaller guarantees fund the cost line. Next, add retention: ESOPs with four-year vesting and a one-year cliff — the poached engineer forfeits unvested value, the mathematics of loyalty — plus a faster promotion track for the top decile. Finally, run the equity checks. The Adams' ratio audit: did the contributor-versus-coaster gap widen justifiably or resentfully? The expectancy audit: is the output-to-payout line of sight visible within the quarterly cycle? The lesson: retention crises are usually structure crises wearing a salary costume. Therefore, the answer that raises everybody's base buys twelve months — and recreates the crisis at a higher price.

The Pay-Compression Case (The Modern Classic)

New graduate hires at ₹12 lakh against three-year engineers at ₹11 lakh — the 2021–24 tech market's signature distortion. Each fix carries examinable costs. First, across-the-board increases: inflationary, unaffordable, and rewarding the underpaid and overpaid alike. Second, selective adjustments for the critical roles: internal-equity grievance from everyone else. Third, structural fixes: new career ladders (splitting title bands so experience differentiates), skill-based pay (certifications and scope, not tenure), and honest communication of the market reality. Therefore, the textbook answer layers all three: adjust the critical 20% immediately, build the ladder within two quarters, and publish the pay philosophy. The concept to name: pay compression inverted the traditional premium for experience. Consequently, its reversal is an organisational-design project, not a payroll adjustment.

Frequently Asked Questions

What are the components of total rewards?

Base pay, dearness-type adjustments, performance-linked incentives, benefits and perquisites (statutory and voluntary), and long-term incentives. The objectives: attract, retain, motivate — with equity, compliance and cost-effectiveness alongside.

Which four laws does the Code on Wages 2019 subsume?

The Payment of Wages Act 1936, the Minimum Wages Act 1948, the Payment of Bonus Act 1965, and the Equal Remuneration Act 1976. Its additions include the national floor wage and the uniform definition of wages.

What is the gratuity formula?

15/26 × last drawn wages × completed years of service, after five years' continuous service for eligibility under the Payment of Gratuity Act 1972.

What are the job evaluation methods?

Ranking, classification or grading, factor comparison, and the point method. The point method — compensable factors weighted by degrees — is the most defensible and widely used.

What is gainsharing?

Group productivity-sharing plans — Scanlon, Rucker and Improshare — distributing measured productivity gains between firm and workforce. Profit sharing and ESOPs complete the group ladder.

What theories anchor compensation design?

Adams' equity theory (fair input-output ratios against referents), Vroom's expectancy theory (credible effort-performance-reward links), and agency theory for executive pay alignment.

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Related reading: HRM Part 2: Recruitment and Selection Methods · HRM Part 4: Performance Appraisal Methods and Biases

HRM Part 6: Industrial Relations and Disputes

Aug 20, 2026

In one line: Industrial relations runs on Dunlop's three actors (workers, employers, the state), the conciliation-to-adjudication dispute ladder, and the Industrial Relations Code 2020's reorganisation - the 51% negotiating union, the 14-day strike notice, and the 300-worker threshold for retrenchment permission.

Industrial relations (IR) covers four things. First, the actors: workers, employers, and the state - and how they interact within a shared ideological frame. Second, the trade union structure and its legal framework under the Trade Unions Act 1926. Third, the dispute-resolution machinery, from negotiation and conciliation up to adjudication by tribunals. Finally, the Industrial Relations Code 2020's reorganisation of this entire landscape. This note covers the full IR file, from theory to the latest reform debate, so that you can answer both objective and descriptive questions with confidence.

Quick Answer: IR studies the relationships between employers, employees and the state. Dunlop's systems model is the theoretical anchor: three actors interacting within contexts, bound by an ideology, producing a web of rules. Meanwhile, disputes escalate through negotiation, conciliation, voluntary arbitration, and adjudication. The IR Code 2020 subsumes the Trade Unions Act 1926, the Standing Orders Act 1946, and the Industrial Disputes Act 1947. Its key changes: the 51% negotiating union, the 14-day strike notice, the 300-worker threshold, and fixed-term employment.

1. IR: The Concept, Actors and Objectives

Industrial relations began as a distinct field of study in the mid-twentieth century, when industrialisation made workplace conflict a matter of public policy rather than private quarrel. Before turning to law and machinery, fix the conceptual base.

• The definitions. IR means the relationships between employers and employees - and their organisations - plus the state's regulation of those relationships. The theoretical anchor is Dunlop's systems model: actors (workers, employers, government) interact within contexts (technology, market, power), bound by a shared ideology, producing a web of rules. The "web of rules" covers everything from wage awards to unwritten workplace norms - a point examiners reward when you elaborate it.
• The objectives. Five: industrial peace and harmony, higher productivity, industrial democracy, protection of the legitimate interests of both sides, and dispute prevention. Moreover, note the modern narrative for long answers - the shift from pluralist conflict frames, which accept competing interests as natural, to HRM-unitarist and partnership frames, which presume aligned interests between employer and employee.
• Why the state matters. The state is not a neutral umpire alone; it legislates, conciliates, and adjudicates. This triple role - regulator, mediator, and adjudicator - explains why every dispute ladder below ends in a statutory forum rather than a purely private settlement.

2. Trade Unions: Structure and Law

Trade unions are IR's primary vehicle for worker voice. Two questions matter: how unions are organised, and how the law protects (and restricts) them.

• The structure ladder. Primary unions at the plant level rise to federations - name the five national centres: INTUC, AITUC, HMS, CITU, and BMS. Then come plant, industry, and national levels of organisation. In addition, know the recognition question: statutory recognition versus voluntary recognition, with the check-off system - where employers deduct union dues directly from wages - serving as a membership-verification device.
• The Trade Unions Act 1926. Registration requires 7+ eligible members, with the registrar holding discretionary refusal grounds (for example, an unsuitable name or an improperly drafted constitution). Then the immunities: Sections 17-18 protect lawful union acts from criminal conspiracy charges and civil suits, including the peaceful picket line. Furthermore, the political fund (Section 16) must remain separate from general union funds - these are the tested provisions.
• The union realities. The critique list for mains answers: multiplicity and inter-union rivalry, outside (political) leadership, politicisation of industrial disputes, declining density amid organised-sector stagnation, and the gig-era representation gap. Pair each critique with a consequence - rivalry, for instance, weakens bargaining and complicates the very recognition question the 2020 Code now tries to answer.

3. The Dispute Machinery

This is the heart of any IR question. Learn the escalation ladder and the statutory anchors together.

• The machinery ladder. The escalation sequence runs: negotiation and collective bargaining (the internal, bilateral stage); then conciliation through the Conciliation Officer and Boards of Conciliation (third-party facilitation that persuades but does not decide); then voluntary arbitration, where parties refer the dispute to a neutral arbitrator whose award binds them; finally, adjudication through labour courts, industrial tribunals, and national tribunals. Remember the logic: each rung adds a third party with progressively greater authority.
• The dispute taxonomy. First, rights disputes - conflicts over the interpretation of existing agreements, standing orders, or awards. In contrast, interest disputes seek entirely new terms, such as higher wages or revised conditions of service. Add the individual-versus-collective split: an individual grievance may become a collective dispute once a union espouses it - a nuance that has decided many borderline cases.
• The weapons. On the worker side, the strike carries legal conditions: advance notice, ballots where required, and prohibitions during the pendency of proceedings, with public-utility and essential-service limits. On the employer side sit lockout, layoff, retrenchment, and closure, each with its own compensation formulae. Retrenchment pays 15 days' average pay per completed year of service; closure pays similarly, with enhanced terms for longer-tenured workmen - these are the numerical MCQs.
• The procedural notes. From the Industrial Disputes Act 1947, hold the essentials: S.2(k) defines "industrial dispute"; S.22-23 fix the conditions under which strikes and lockouts become illegal in public utilities; S.25F sets the conditions - one month's notice or wages in lieu, 15 days' average pay per completed year, and prior permission where required - for valid retrenchment.

4. The Codes Era: The IR Code 2020

The Code on Industrial Relations 2020 replaces three twentieth-century statutes with a single framework designed for flexibility and faster dispute settlement. Learn the subsumption list first, then the change list.

• The subsumption. The Code on Industrial Relations 2020 is enacted, with rules notification proceeding in stages - always add the status caveat. It subsumes three laws: the Trade Unions Act 1926, the Industrial Employment (Standing Orders) Act 1946, and the Industrial Disputes Act 1947 - the subsumption MCQ appears almost every cycle.
• The key changes. First, the negotiating union/council: sole bargaining recognition at 51% membership, or, failing that, a negotiating council formed from unions with 20%+ membership combined. Then, the strike-notice tightening: a 14-day written notice, with strikes prohibited during conciliation and for 7 days after its conclusion - and these provisions now extend to all establishments, not merely public utilities. Next, the standing-orders and retrenchment-permission threshold raise: from 100 to 300 workers, meaning establishments up to 300 can alter service conditions or effect retrenchment and closure without prior government permission. Furthermore, fixed-term employment is formally recognised, with equal benefits and gratuity computed on the basis of length of service as if continuous. Add the two-member Industrial Tribunal flexibility, the re-skilling fund credited with 15 days' wages for each retrenched worker, and the mandatory Grievance Redressal Committee at establishments with 20+ workers - this is the change list every paper tests.
• The debate. The flexibility-versus-security critique runs both ways: the 300-threshold liberalisation may ease hiring but arguably dilutes job security and weakens the "prior permission" check; the 51% negotiating-union hurdle may stabilise bargaining but could marginalise small and independent unions. Argue both flanks in long answers, and close with a balanced judgement.

5. Workers' Participation and Collective Bargaining

Participation and bargaining are IR's constructive side - the mechanisms by which conflict is channelled into agreement rather than confrontation.

• Participation schemes. The participative-management ladder ascends as follows: works committees (the ID Act's consultative bodies for establishments with 100+ workers), joint management councils, shop-floor and plant-level councils, and, at the apex, board-level worker representation. However, India's experience is best summarised as weak statutory consultative bodies - committees often meet irregularly, and their advisory status limits influence. This is the standard evaluation line for mains.
• Collective bargaining levels. Bargaining operates at the plant or firm level, the industry level, and the national level. The process runs in three stages: preparation (mandate, data, priorities); then negotiation itself (ground rules and the "zone of agreement" - the overlap between each side's minimum acceptable and maximum aspirational outcomes); finally, agreement administration and grievance handling. Meanwhile, the determinants of bargaining power are union density, the substitutability of labour, and broader economic conditions. The honest assessment: Indian bargaining concentrates at the plant level, with thin coverage beyond the organised sector - a coverage gap the 2020 Code's negotiating-union design tries to address.
• The modern layer. The frontier paragraph for contemporary answers: gig-platform bargaining experiments, the European works-council comparison (statutory information-and-consultation bodies in transnational firms), and partnership-model cases where unions and management jointly pursue competitiveness.

6. How Exams Probe This Topic

Knowing the pattern saves preparation time. Here is how UPSC, UGC-NET, and state services questions typically approach IR.

• MCQs: the IR Code's subsumed acts; the 51% negotiating-union rule; retrenchment compensation figures; strike-notice periods; the five national federations; the TU Act's immunities; Dunlop's actors and contexts.
• Short answers: the dispute-machinery escalation; rights versus interest disputes; the 2020 Code's key changes; levels of collective bargaining; the case for and against fixed-term employment.
• Cases and long answers: analysing a strike's legality step by step; auditing a retrenchment for S.25F (or Code-equivalent) compliance; designing a workers' participation structure for a described plant, justifying each tier.

7. Quick Revision: One-Glance Facts

• Frame. Dunlop's three actors + web of rules; objectives: peace, productivity, democracy, protection, prevention.
• Machinery. Negotiation, conciliation, voluntary arbitration, adjudication (labour court, industrial tribunal, national tribunal).
• Code 2020. Subsumes TU Act 1926 + Standing Orders Act 1946 + ID Act 1947; 51% negotiating union (council at 20%+); 14-day strike notice; 300-worker threshold; fixed-term employment; re-skilling fund at 15 days' wages.
• Numbers. Retrenchment: 15 days' average pay per completed year; Grievance Redressal Committee at 20+ workers; Works Committee at 100+ workers; TU registration at 7+ members.

Conclusion. IR answers run on the machinery ladder and the Code-2020 change list: the actors' triangle, the conciliation-to-adjudication escalation, and the reorganised rules - with the flexibility-versus-security debate as the analytical layer that elevates an answer from good to excellent. Therefore, fix the subsumption lists, the 51%-and-14-day rules, and the compensation formulae. With those anchors, any HRM paper's dispute-machinery question works for you.

The Code-2020 Transition File (The Implementation Layer)

What examiners want goes beyond the statutory text. First, the rules-notification status: the four labour codes were enacted in 2019-20, but final rules and enforcement have proceeded in stages through 2022-25 - always verify the current status in exam week. Second, the negotiating union in practice: the 51% threshold and the 20%-plus council remain largely untested, so the first recognition cycles are the watch-point. Third, fixed-term employment's adoption: uptake is growing in seasonal and manufacturing roles, with equal benefits as the statutory safeguard - however, the debate continues over whether it formalises flexibility or informalises permanence. Finally, the industrial-peace question: the long-run decline in strike man-days predates the Code. Therefore, attributing that decline to the reform would be causally careless - and saying so, with reasoning, marks a careful answer. The honest frame: the Code's text and its practice still diverge. Implementation, not enactment, is the decade's story.

The Gig-and-Platform Extension (The Frontier Layer)

Industrial relations' new terrain has three files. First, the Code's gap: the IR Code presumes an employment relationship, which the platform worker typically lacks. Consequently, such workers fall back on the Code on Social Security 2020's aggregator-levy framework and its roughly 1-2% turnover contribution funding welfare schemes. Second, the state innovations: Rajasthan's 2021 gig workers Act - India's first, establishing a welfare board and a transaction cess - and Karnataka's subsequent legislation show the federal structure enabling policy experimentation. Third, the new bargaining question: collective voice is forming through platform-specific unions and app-based coordination. However, against algorithmic management - where scheduling and discipline are code rather than supervisors - the traditional strike's logic needs reinvention. The extension's thesis: the next decade's IR debate will be written on the platforms. Therefore, answers connecting the 1926 Act's framework to the 2025 gig-union reality read as genuinely contemporary.

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Frequently Asked Questions

What is Dunlop's model of industrial relations?

Three actors - workers, employers, and government - interact within the contexts of technology, market, and power distribution, bound by a shared ideology, and together produce a web of rules. It remains IR's standard theoretical anchor in exams and textbooks alike.

Which laws does the Industrial Relations Code 2020 subsume?

Three: the Trade Unions Act 1926, the Industrial Employment (Standing Orders) Act 1946, and the Industrial Disputes Act 1947. Rules notification has proceeded in stages - verify the status at exam time.

What is the negotiating-union rule under the Code?

A union with 51% membership becomes the sole bargaining agent. Alternatively, a negotiating council can form from unions whose combined membership is 20%-plus.

What is the retrenchment compensation formula?

15 days' average pay per completed year of service, alongside the Section 25F conditions (notice or wages in lieu, and prior permission where applicable). The Code's re-skilling fund adds a further 15 days' wages for each retrenched worker.

How do strikes become illegal?

Without the required notice - now 14 days under the Code - or during conciliation pendency and for 7 days after its conclusion, or against public-utility and essential-service prohibitions (S.22-23 of the ID Act).

How does IR reach gig workers?

Indirectly: the IR Code presumes an employment relationship, so platform workers rely on the Social Security Code 2020's aggregator levy - plus state-level laws like Rajasthan's 2021 gig workers Act and Karnataka's subsequent legislation.

HRM Part 7: The Factories Act Link to HR Compliance

Aug 20, 2026

In one line: The Factories Act 1948 is HR-compliance's statutory backbone: the 10/20 factory definition, the health-safety-welfare section ladders (S.11-50), the 48-hour week, and the 30-150-250-500-1000 thresholds — now subsumed by the OSH Code 2020.

The Factories Act 1948 is HR-compliance's statutory backbone. It covers health, safety, welfare, working hours and leave, all keyed to the "factory" definition and worker-count thresholds. Moreover, every HRM paper tests its provisions, and every HR role operates inside them — from factory licensing and safety-committee formation to overtime computation and welfare-officer appointments. This note covers the complete compliance file: the definitions that decide when the Act applies, the section-wise provisions that anchor every MCQ, the thresholds that decide what a factory must build and whom it must hire, and the 2020s reform layer that every current answer must acknowledge.

Quick Answer: The Act defines a factory as premises with 10 or more workers using power (20 without power). Health sits in Sections 11-20, safety in 21-41, welfare in 42-50. The threshold ladder: creche at 30 women, first-aid at 150, canteen at 250, welfare officer at 500, safety officer at 1,000. Meanwhile, hours run 48/week and 9/day, with overtime at twice the ordinary rate. Finally, the OSH Code 2020 subsumes the Act — implementation pending.

1. The Act's Scope and Definitions

• The card. The Factories Act 1948 consolidated the 1881-1891-1911-1922 lineage of Indian factory legislation, replacing the 1934 Act with a modern instrument aligned to the ILO's post-war standards. Furthermore, its constitutional anchor sits in the Directive Principles: Articles 39(e)-(f) and 42, which direct the State to secure humane conditions of work and protect workers' health and dignity. Whenever a long answer asks for the Act's philosophical basis, these Articles are the opening line.
• The definitions. First, a factory (S.2(m)) means premises where 10 or more workers with power — or 20 without power — are employed on any day of the preceding 12 months in a manufacturing process. Note the "any day" formulation: one peak-season day beyond the threshold pulls the entire premises inside the Act. Second, the manufacturing process (S.2(k)) is deliberately broad: making, altering, repairing, ornamenting, finishing, packing, oiling, washing, breaking up or demolishing, even printing-type processes and generating or transforming power. Third, a worker (S.2(l)) covers a person employed directly or through an agency, with or without the employer's knowledge — an inclusive reading that matters in contract-labour audits. Finally, the occupier (S.2(n)) is the person with ultimate control over the factory's affairs, fixed with strict liability by the case law (notably the Supreme Court's line of occupier-liability decisions). The occupier is the compliance face of the enterprise; penalties attach to that person first.
• The approval-first rule. A factory cannot simply open its gates. Its site plan, building drawings and machinery layout need prior approval and licensing by the state's Chief Inspector before operations commence (S.6). In addition, the state government holds wide rule-making and exemption powers — which is why state-level Factories Rules, not the central Act alone, govern day-to-day compliance formats such as register maintenance and notice-of-accident filings.

2. Health Provisions (Sections 11-20)

The section-number to provision map is the MCQ staple — ten sections, ten memorisable hooks. First, cleanliness (S.11) sets daily-cleaning, effective drainage and whitewashing cycles (whitewash every 14 months in the standard framing). Then come disposal of wastes and effluents (12), which requires sanitary arrangements to the state rules' standard; ventilation and temperature (13), which requires reasonable comfort and the removal of vitiated air; and dust and fume (14), which requires exhaust appliances where dust is generated. Next, artificial humidification (15) — relevant to textile mills, with a prescribed water-quality standard; overcrowding (16), which caps workers per unit of cubic space (roughly 350 cubic feet per worker in the standard reading, excluding space above 14 feet from the floor); and lighting (17), which requires sufficient and suitable light, protected against glare and reflection. Finally, drinking water (18 — cooling in hot season, marked points within 6 metres of every workstation, and separate points where over 250 workers are employed), latrines and urinals (19 — separate accommodation for either sex, screened and accessible), and spittoons (20 — sufficient number, with penalties for spitting outside them).

A memory aid: the health ladder runs roughly from the building outward — clean it (11), drain it (12), air it (13-14), humidify it (15), decongest it (16), light it (17), water it (18), then provide the conveniences (19-20).

3. Safety Provisions (Sections 21-41)

The safety battery is the Act's longest chapter, and the case-method exam's favourite source of violations. Machine guarding (21-26) carries the fencing rules for moving parts, Prime-mover and transmission-machinery guards, and the prohibitions on adolescents working at dangerous machines (S.23). Then, revolving machinery's speed limits must be visibly stated (25 — the centrifugal machines rule), self-acting machines need clearance space (26), and the casing of new machinery is required at the point of purchase (28). The prohibition on women near cotton openers (34) is a dated but still-asked provision. Next, eye protection (35) covers grinding, welding and similar high-risk processes; restrictions on dangerous operations (S.41 and the rules) allow the state to prohibit employment in hazardous processes; and precautions against dangerous fumes (36) contain the confined-space entry rule — no person may enter any chamber or confined space where dangerous fumes are likely present without a certified-air test and standby arrangements. Finally, explosive or inflammable materials demand segregation, enclosure and precautions (37 and 37A — the latter on precautions against dangerous fumes in grinding processes), and safety officers' appointment (40B) is triggered at the 1,000-worker threshold, where the process involves hazardous operations.

Two further hooks: hoists and lifts (28-29) need periodic examination and a safe working load marked; and floors, stairs and means of access (32) must be sound and free from obstruction — the mundane provision that examiners love because it is easy to violate in a case study.

4. Welfare Provisions (Sections 42-50)

The welfare chapter is where the worker-count thresholds live — the exam's favourite MCQs and the compliance audit's triggers. Washing facilities come first (42 — adequate and suitable, separate for either sex). Then storing and drying clothing (43), seating arrangements for workers standing at work (43's companion), and first aid (44 — one box per 150 workers, each box in charge of a trained responsible person, with a dispensary, room and medical staff where 500 or more workers are employed). Next, the canteen (46 — mandatory at 250 or more workers) and shelters, restrooms and lunchrooms (47 — 150 or more workers; the lunchroom exception applies where workers take meals at the canteen). Then the creche (48 — where 30 or more women workers are employed, a suitable room for children under six years, in the women's use and under trained supervision). Finally, welfare officers (49 — 500 or more workers, with prescribed qualifications) and the S.50 power enabling the state to extend further welfare rules.

Memorise the 150-250-500-30 combinations as a ladder, not a list: as headcount rises, the statute walks the factory up from a first-aid box to a canteen, a dispensary, and finally a full-time welfare officer — with the women-specific creche trigger sitting independently at 30.

5. Working Hours, Leave and the Enforcement Machinery

• Hours and OT. Adults work 48 hours a week and 9 a day (S.51-54), within a spread-over of 10½ hours (extendable to 12 by rule) and prescribed rest intervals. Weekly holidays (52) require compensatory arrangements when work occurs on the rest day. Overtime runs at twice the ordinary rate of wages (59), and the register-of-overtime compliance flows from the state rules. Meanwhile, women's statutory window is 6 a.m.-7 p.m. (S.66), with state-wise relaxations subject to safeguards — see the modern-layer amendment discussed below. Furthermore, adolescents' and children's employment now reads through the Child Labour (Prohibition and Regulation) Amendment Act 2016: complete prohibition below 14 years, non-hazardous work permitted for 14-18-year-olds with conditions — the interface HR answers must mention whenever "adolescent" appears in a question.
• Leave. Annual leave with wages (S.79) accrues at one day per 20 days worked for adults, and one per 15 for children. However, eligibility needs 240 days' service in the calendar year — the calculation MCQ: count the days worked, apply the divisor, and check carry-forward limits (capped at 30 days for adults, 40 for children).
• Enforcement. The Chief Inspector and Inspectors (S.8-11) hold entry, examination, seizure and inquiry powers, and can require plans and registers. In addition, safety committees (41G) provide workers' participation in safety management — mandated at hazardous-process factories. Penalties (S.92-93) fix the occupier's and manager's liability for general contraventions, with imprisonment and fine enhanced on repetition (S.94-96 for specific offences). Finally, the reform layer: the Occupational Safety, Health and Working Conditions Code 2020 subsumes the Factories Act in a 13-law consolidation — the factory threshold itself changed (10/20 restructured around power use, and a new "employee" count for certain obligations), licence renewals made long-cycle, and the two-women night-work gate introduced with consent and safeguards. Implementation remains pending with state rules under frame; verify status before the exam.

6. How Exams Probe This Topic

• MCQs: the 10-and-20 factory thresholds; the section-to-provision matches (18-water, 19-latrine, 35-eye protection, 40B-safety officer, 44-first-aid, 46-canteen, 47-shelters, 48-creche, 49-welfare officer); the 48-hour week and 9-hour day; the spread-over; the OT double-rate; the 240-day leave rule with the 20-day divisor.
• Short answers: the health-safety-welfare trio's scheme; the inspectorate's powers; the safety-committee role at hazardous factories; the OSH Code's subsumption and its changed thresholds.
• Cases: a compliance audit of a described factory, identifying missing provisions with sections; the creche and welfare design at stated headcounts; an hours-and-overtime computation from shift data.

7. Quick Revision: One-Glance Facts

• Scope. 10 with power / 20 without; manufacturing process broadly defined (S.2(k)); the occupier's strict liability.
• Thresholds. First-aid 150; shelters 150; canteen 250; creche 30 women; welfare officer 500; safety officer 1,000.
• Hours. 48/week, 9/day, 10½-hour spread-over; OT 2x; leave one day per 20 (adults) after 240 days.
• Enforcement. Inspectors' entry powers (S.8-11); safety committees (41G); penalties (S.92 onwards).
• Reform. The OSH Code 2020 subsumes it — implementation pending, verify.

Conclusion. The Factories Act is threshold and section recall: the 10-20 definition, the health-safety-welfare ladders, and the 30-150-250-500-1000 numbers. Therefore, pair the recall with the current layer — the OSH Code's subsumption and the women's night-work relaxations — and HR-compliance questions in any management or law paper are fully answered from this note.

Practice Corner: Five Threshold Checks (with Answers)

1. Creche required at how many women workers? — 30+.
2. Canteen at? — 250+ workers.
3. Welfare officer at? — 500+.
4. Annual leave accrues at? — One day per 20 days worked (adults), after 240 days' service.
5. Overtime is paid at? — Twice the ordinary rate.

Score five out of five before moving on. If any answer needed a second look, re-read the welfare chapter — thresholds are pure recall, and pure recall is free marks.

The Compliance-Audit Case Method

Picture the classic case: a factory with 400 workers, 40 women, 12-hour shifts, no canteen, a first-aid box with no attendant. Now walk the statute in four moves — welfare, hours, safety, liability. First, the canteen is missing — required at 250+ (S.46). Then, the creche is missing, since 40 women exceed 30 (S.48). Meanwhile, the 12-hour shifts breach the 9-hour cap and the 48-hour week (S.51-54), attracting overtime at double rate and Section 92 penalties. In addition, first aid needs the prescribed scale — more than one box at 400 workers — with a trained attendant (S.44 and the rules). Finally, the occupier — not merely the manager — bears the strict liability. Close with the OSH Code 2020's subsumption to show reform awareness. Every threshold in this Act exists to be caught in exactly this audit format — learn them as triggers, not as a list.

The Compliance-Audit Case (Fully Worked)

Take a factory with 420 workers (38 women), two 10-hour shifts, a canteen, no creche, one first-aid box — and an unguarded lathe. First, hours (S.51-54): the 10-hour shifts breach both the 9-hour daily and 48-hour weekly caps. Consequently, the excess hour daily per worker becomes overtime at twice the ordinary rate, and repeated breaches attract S.92's penalties — the occupier named first. Then, welfare (S.42-50): 420 workers require nearly three first-aid boxes on S.44's one-per-150 scale, each with a trained attendant; a single box is a plain violation. The canteen stands compliant at 250+; however, the creche at 38 women (30+) is missing — S.48 violated. The welfare officer (500+) is not yet triggered — say so explicitly; examiners reward the negative finding as much as the positive. Next, safety (S.21): the unguarded lathe is the classic fencing violation, and any accident escalates to S.92-92A liability plus the occupier's personal exposure under the strict-liability doctrine. Finally, the reform layer: note the OSH Code 2020's subsumption, with its pending rules and changed thresholds. In short, the case method is the whole Act in one exercise: thresholds, hours, welfare, safety, penalties, reform — in that order, every time.

The Women's Hours Amendment (The Modern Layer)

The 2024-25 rule amendments moved to relax the night-work prohibitions for women in factories. Consequently, state-wise notifications now permit 6 p.m.-6 a.m. work with prescribed safeguards: written consent, transport between residence and factory, adequate lighting, security, canteen and restroom facilities, and grievance mechanisms. Several states have aligned their Factories Rules accordingly, and the OSH Code 2020 carries a similar two-women-consent gate for night work. The exam angle is twofold. First, the constitutional frame — Articles 14, 15 and 42, protective legislation versus equality's modern reading: the older paternalist bar is increasingly seen as excluding women from night-shift employment rather than protecting them. Second, the labour-economics reading: electronics, pharma and global-capability-centre shift systems — plus the competitiveness argument in states courting manufacturing investment — drove the change. Therefore, note it as a live reform layer with safeguard conditions, and any Factories Act answer carries this decade's fingerprint rather than the 1948 text alone.

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Frequently Asked Questions

What is a "factory" under the Factories Act?

Premises where 10 or more workers with power — or 20 without power — are employed on any day in the preceding 12 months in a manufacturing process (S.2(m)). The manufacturing process itself is defined broadly (S.2(k)), covering making, altering, repairing, packing and printing-type processes.

What are the welfare thresholds every HR paper tests?

Creche at 30 women workers, first-aid at 150, canteen at 250, welfare officer at 500, and safety officer at 1,000. Learn them as a ladder — they anchor nearly every compliance MCQ and every case study.

What are the working-hour limits?

Adults: 48 hours a week and 9 a day (S.51-54), within a prescribed spread-over. Overtime is paid at twice the ordinary rate. Women's standard window is 6 a.m.-7 p.m., now relaxable state-wise to 6 a.m.-6 a.m. night work with safeguards.

How does annual leave accrue?

One day per 20 days worked for adults (one per 15 for children), after 240 days' service in the calendar year (S.79). Carry-forward is capped at 30 days for adults.

Who is the "occupier," and why does it matter?

The person with ultimate control over the factory's affairs (S.2(n)). It matters because penalties and strict liability attach to the occupier first — the standard exam point in any liability-based case question.

What is the OSH Code 2020's effect on the Act?

It subsumes the Factories Act in a 13-law consolidation with changed thresholds, long-cycle licences, and a consent-based night-work gate for women. However, implementation is pending with state rules — always verify the current status before the exam.

HRM Part 8: HR Analytics and the New World of Work

Aug 20, 2026

In one line: HR analytics applies data to people decisions through a four-rung maturity ladder (descriptive-diagnostic-predictive-prescriptive), metric families from attrition to eNPS, and Google's Project Oxygen as the canonical case - bounded by DPDP consent and bias audits.

The HRM series closes with its frontier. First comes the analytics maturity ladder. Then the metric families, from turnover to engagement. Next, the evidence-based-HR logic. Finally, the new work layer: hybrid, gig, and AI in HR. This note covers the modern file - increasingly the differentiator in NET and MBA papers.

Quick Answer: HR analytics is the data-driven approach to people decisions, climbing a four-rung ladder: descriptive (what happened), diagnostic (why), predictive (what will), prescriptive (what to do). The metric families span talent acquisition, retention, productivity, engagement and compensation. Moreover, evidence-based HR combines data, research, stakeholders and judgment. Indeed, the new work layer covers hybrid models, the gig economy under the Code on Social Security 2020, and AI in HR. Finally, the limits: DPDP consent, algorithmic bias audits, and Goodhart's law.

Table of Contents

1. HR Analytics: The Concept and the Maturity Ladder
2. The Metric Families
3. Evidence-Based HR and Predictive Use-Cases
4. The New World of Work
5. Ethics, Privacy and the Limits
6. How Exams Probe This Topic
7. Quick Revision: One-Glance Facts

1. HR Analytics: The Concept and the Maturity Ladder

• The definition. HR analytics is the data-driven approach to people decisions - applying statistics and modelling to HR data, progressing from descriptive to prescriptive. Note the distinction: HRIS is the transactional system, and HR metrics are the raw numbers.
• The maturity ladder (the standard model). Four rungs, in order. First, descriptive - what happened, via dashboards. Second, diagnostic - why it happened, via correlations and drill-downs. Third, predictive - what will happen, via attrition and performance models. Finally, prescriptive - what to do, via intervention recommendations. Reproduce this ladder in any long answer.

2. The Metric Families

• Talent acquisition: time-to-fill, cost-per-hire, quality of hire (performance at 6-12 months), offer-acceptance, and source-effectiveness.
• Retention: the attrition/turnover rate (separations over average headcount). Cut it further: voluntary versus involuntary, and regretted versus non-regretted. In addition, track retention rate and the early-attrition (90-day) signal.
• Productivity and performance: revenue per employee, human capital ROI (revenue minus non-labour costs, over labour costs), and performance distribution.
• Engagement and development: engagement survey scores, the eNPS (employee net-promoter), training ROI (Part 3's Kirkpatrick results), internal-mobility and promotion rates, and the absenteeism rate.
• Compensation: compa-ratio (pay over midpoint), pay-equity gaps, and benefits uptake. Learn the family list with two or three formulas each.

3. Evidence-Based HR and Predictive Use-Cases

• The movement. Evidence-based HR (the Briner-Rousseau lineage) draws decisions from four sources: best available data, research evidence, stakeholder input, and practitioner judgment. Therefore, it stands against intuition and fad - the best-practice skepticism.
• The canonical use-cases. First, flight-risk prediction (supervised models on tenure, engagement and compensation data). Second, workforce planning (demand-supply forecasts). Third, diversity analytics (hiring and promotion funnel gaps). Fourth, sentiment analysis (survey and pulse text mining). Finally, skills-gap mapping for reskilling - the AI era's hottest application.
• The famous case-layer. Google's Project Oxygen (manager-behaviour analytics) and its People Analytics team are the canonical citations. Meanwhile, the Indian IT sector's attrition modelling serves as the domestic example.

4. The New World of Work

• Hybrid and remote. The design choices run from structured-hybrid to remote-first models. However, the outcomes debate continues: productivity evidence is mixed, and proximity bias is a real risk. Indeed, policies keep evolving through the 2022-25 return-to-office waves.
• The gig and platform layer. The Code on Social Security 2020 recognises gig workers (Part 5's statutory link). The workforce numbers sit around 2.3 crore by NITI's estimate. Consequently, the HR challenge becomes engagement and belonging without employment.
• AI in HR. Recruitment-AI covers resume screening and chatbots. Then come workforce-productivity tools, and the augmentation-versus-displacement framing (job design's reconfiguration). In addition, note skills-first hiring and degree inflation's reversal.
• The employee-experience frame. Three current themes: the moments-that-matter journey design (candidate to alumnus); wellbeing and mental-health programmes' mainstreaming; and DEI's institutionalisation.

5. Ethics, Privacy and the Limits

• The privacy layer. Employee data needs consent and purpose. Therefore, the DPDP Act 2023 applies to HR data - balancing consent, necessity and legitimate use. Indeed, algorithmic accountability develops abroad: the EU AI Act classifies employment AI as high-risk, while India's debates emerge.
• The bias risk. Historical data carries bias - the Amazon screening-tool case is the canonical cautionary tale. Consequently, audits and explainability requirements follow.
• The analytical limits. Correlation is not causation. Moreover, Goodhart's law looms: metrics gamed become useless. Meanwhile, preserve human judgment - the balanced close every ethics-flavoured answer needs.

6. How Exams Probe This Topic

• MCQs: the maturity rung order; metric definitions (attrition rate, compa-ratio, eNPS, HC-ROI); evidence-based HR's four sources; Project Oxygen; DPDP's HR interface.
• Short answers: HRIS versus metrics versus analytics; predictive use-cases; algorithmic bias and remedies.
• Cases: design an analytics function for a described firm (the maturity roadmap); an attrition-prediction programme with privacy safeguards.

7. Quick Revision: One-Glance Facts

• Ladder. Descriptive, diagnostic, predictive, prescriptive.
• Metrics. Attrition (separations over average headcount), time-to-fill, compa-ratio, eNPS, HC-ROI, revenue per employee.
• Evidence-based HR. Data + research + stakeholders + judgment; Google's Oxygen.
• New work. Hybrid, gig (the Code 2020), AI in HR, EX and DEI.
• Limits. DPDP consent; bias audits; Goodhart's law.

Conclusion. HR analytics is the field's evidence-based turn: the maturity ladder, the metric families with their formulas, and the predictive use-cases - bounded by consent and bias safeguards. Therefore, hold the ladder, the formulas, and the two canonical cases (Oxygen and the Amazon caution). With those, the frontier questions are already answered.

Practice Corner: Five Metric Checks (with Answers)

1. The maturity ladder's four rungs? - Descriptive, diagnostic, predictive, prescriptive.
2. The attrition rate formula? - Separations over average headcount (cut further: regretted vs non-regretted).
3. What is eNPS? - The employee net-promoter score: promoters minus detractors on the "would you recommend" question.
4. The famous manager-behaviour study? - Google's Project Oxygen.
5. The ethical anchors? - DPDP consent, algorithmic bias audits, Goodhart's law (gamed metrics stop working).

The Case Lens: An Attrition-Prediction Programme (With Safeguards)

A 2,000-employee IT firm wants to predict flight risk. First, assemble the ethically available predictors: tenure, compensation ratio, promotion velocity, engagement pulses, and manager-change events - never health or personal-life data. Then build the model on exit-tagged history. Next, output team-level risk aggregates rather than individual blacklists. Route the outputs to retention interventions - stay-conversations and career-path fixes - rather than pre-emptive exits. Meanwhile, audit quarterly for bias: are single managers' teams or returning-mothers' cohorts flagged disproportionately? Finally, disclose the programme's existence per the DPDP consent regime. The unsafeguarded version - an individual score shared with line managers - is the case's designed failure mode. Therefore, every analytics case rewards the same pairing: the technical pipeline plus the governance wrapper around it.

The Analytics Maturity Case (Fully Worked)

A retail chain with 200 stores wants "HR analytics" - walk the ladder honestly. Descriptive (month one): the dashboard - attrition by store and tenure, hiring-funnel conversion, overtime and absenteeism heatmaps. However, the risk here is vanity metrics: charts nobody acts on. Diagnostic (quarter one): why does north-region attrition run six points high? Correlate exit-interview themes, manager tenure, pay compa-ratios, and commute profiles. Consequently, the drill-down finds the manager effect (stores under first-year managers churn higher), the pay effect (the 0.9 compa-ratio cluster), and their interaction. Predictive (quarter two): the flight-risk model on the diagnostic's variables, validated at, say, 0.75 AUC - honest, useful, not magic. The output feeds retention interventions, not blacklists. Prescriptive (quarter three): the intervention engine - stay-conversations triggered at risk thresholds, compa-ratio fixes for the flagged-and-underpaid, manager coaching routed by the manager effect. Finally, the A/B measurement (treated versus matched control stores) closes the evidence loop. The examinable spine: each rung's decision value, not its chart value - plus the governance wrapper of consent, bias audit and DPDP compliance.

The Goodhart-and-Bias Layer (The Two Failure Modes)

Goodhart's law in HR: when the metric becomes the target, the metric dies. For example, the recruitment team measured on time-to-fill hires fast-and-wrong. Similarly, the L&D team measured on training hours schedules hours, not learning. Indeed, the retention team measured on attrition reclassifies exits as voluntary transfers. The countermeasure: paired metrics (speed with quality, volume with validity) and a periodic audit of each metric's gaming potential - a five-minute exercise most analytics functions never run. Algorithmic bias in HR: the Amazon screening case - a model trained on male-dominated history learned to penalise women's resumes - is the canonical citation. Consequently, the modern duty has three parts: the bias audit (does the flag rate differ by protected group?), the explainability standard (can a rejected candidate be told why?), and human review for adverse decisions. Two failure modes, two disciplines. Therefore, naming both in any analytics answer marks the difference between a technician's answer and a manager's.

Read next: DPDP Act 2023: India's Privacy Law in Force, Exam-Ready Notes

Frequently Asked Questions

What is the HR analytics maturity ladder?

Four rungs: descriptive (what happened - dashboards), diagnostic (why - drill-downs), predictive (what will - attrition models), and prescriptive (what to do - intervention recommendations).

What is the attrition rate formula?

Separations over average headcount. For insight, cut it voluntary-versus-involuntary and regretted-versus-non-regretted, and watch the 90-day early-attrition signal.

What are evidence-based HR's four sources?

Best available data, research evidence, stakeholder input, and practitioner judgment - the Briner-Rousseau frame against intuition and fad.

What is Google's Project Oxygen?

Google's manager-behaviour analytics study - the canonical HR-analytics citation, run by its People Analytics team.

What are the ethics limits of HR analytics?

Three: DPDP Act consent for employee data, algorithmic bias audits (the Amazon screening case), and Goodhart's law - metrics gamed become useless.

More free series: hmmnm.in