Specialist Option 4: Organisational Psychology A2

Organisational psychology is not only a toolkit for extracting productivity. A defensible application balances performance/safety with autonomy, fairness, privacy, health and unintended stakeholder effects.

Match conclusions to outcomes: intrinsic motivation/task performance, evaluated leader competence, vigilance detection, recorded safety events, or sabotage justification. None alone proves a universal workplace policy or actual long-term organisational success.

Do not move directly from a significant result to policy. First identify the measured outcome, causal strength, worker/manager trade-offs, possible gaming or harm, and whether the result survives in the intended workplace over time.

McClelland's three needs are not a hierarchy: people can show all three in different strengths. Both theories risk deterministic or culturally narrow predictions if they ignore choice and situation; qualitative individual evidence may reveal personal priorities, while large-sample measures seek general patterns.

Because the links multiply, a near-zero link can collapse motivation even when the others are strong. Expectancy is not instrumentality: the first connects effort to performance; the second connects performance to outcome. VIE is cognitively broad across three perceptions, but still reduces a changing social workplace to an individual decision model.

The study supports functional meaning, not ‘money always motivates’ or ‘money always undermines motivation’. Random allocation strengthens causal inference for the brief task, but student volunteers, self-report mediators, a small gift card and timed anagrams limit generalisation to employees, cultures, complex work and long-term reward systems.

Universalist theories favour individual/nature explanations; behavioural and adaptive accounts make nurture and situation more visible. These are theoretical models, not proof of one universal best style, and their application must be checked across tasks, cultures, power relations and outcomes.

The four controlled narratives isolate style stereotype and leader sex, strengthening causal comparison. They do not show that women are worse leaders or that one style works best everywhere: participants rated a fictional, gender-neutral hospital scenario, not real behaviour or performance. Student/self-report data and Spanish context constrain generalisation.

Standardised LPI items support repeatable numeric comparison over practices, observers and time, but self/observer ratings can differ through interpretation, limited opportunity to observe, social desirability or workplace politics. A score reports perceived frequency, not objective leadership success. Kelley labels are also nomothetic simplifications: role, leader, culture and situation may shift behaviour, and the model does not establish whether styles are innate, learned or fixed.

These are diagnostic models, not a fixed script. Groups can revisit stages; one person can fill several roles; nine people are not required. Cohesion supports coordination but becomes risky when it suppresses disconfirming evidence—so role balance must be paired with decision checks.

Random allocation, a no-monitoring control and standardised signals/equipment support causal inference and reliability. But a short lab vigilance task with student volunteers does not prove that surveillance improves complex work, long-term productivity or welfare. Monitoring may change autonomy, stress, privacy and behaviour; fewer errors here must not substitute for an ethical workplace evaluation.

A target's anxiety or sensitivity may shape reporting or vulnerability but does not cause or justify aggression. Diagnose repeated acts, duration, power and system response before labelling ordinary disagreement as bullying. Conflict modes are situational tools; ethical management must stop harm, protect reporting and correct organisational conditions rather than ask the target simply to accommodate or avoid.

Repeated measures show within-worker change in a real setting and interviews add meaning, but office change is not random: novelty, management, task and time may co-vary, repeated questionnaires can sensitise responses, and attrition can bias later waves. Treat layout as a bundle of conditions, not a deterministic cause.

The survey studies real workers and quantitative odds permit comparison, but self-reported sleep/errors may involve recall, social desirability and reporting differences. Shift type was not randomly assigned; workload, health, childcare and self-selection can confound the association. Therefore Gold supports a useful risk relationship, not a deterministic causal claim for every worker or job.

Recorded data can look objective, but fear of blame/job loss, memory, inconsistent definitions and under-reporting make it incomplete. Low counts may mean concealment, not safety. Combine records with observation/interviews and treat individual error as one link in a system; prevention should make the safe action easy, visible and reinforced while removing hazards at source.

Removing bad conditions is not the same as creating fulfilling work, and adding tasks is not automatically enrichment. Diagnose the missing factor/characteristic, co-design the change, provide training and check satisfaction, safety and performance over time and across cultures.

Standard wording/scoring supports reliability and quantitative comparison, not objectivity of the underlying self-report. Social desirability, fear about employment, acquiescence, ambiguous/complex QWL vocabulary, fatigue and cultural interpretation can reduce validity. Use anonymity, accessible piloting, facet-level reporting and optional interviews/open comments; never treat one total as the employee's complete experience.

Questionnaire acceptance of a reason is not observed sabotage, and an absence category is an interpretation—not proof of motive. Use records, context and employee voice before intervention. Reducing sabotage or withdrawal should address fair pay, voice, job design, leadership and security as well as proportionate controls; blanket surveillance may suppress reporting and deepen the grievance.