Extended self-determination and job demands models explain police well-being

It is a sobering irony of policing that the greatest threat to an officer’s health is usually not the armed offender or the traffic stop gone wrong, but the slow accumulation of organisational strain. More than three decades ago, a sergeant writing in the FBI Law Enforcement Bulletin warned that “it probably won’t be a bullet that strikes an officer down, but the effects of chronic stress.” A major new study of nearly 4,000 Australian police personnel, published in the journal Current Psychology, now provides the most comprehensive evidence to date for why that warning holds true — and, crucially, for what police organisations can do about it. The research, led by Herbert W. Marsh, Richard M. Ryan, Rhonda G. Craven, Theresa Dicke, Jiesi Guo and Philip D. Parker of the Institute for Positive Psychology and Education at Australian Catholic University, integrates two of the most influential frameworks in occupational psychology into a single model of how command-and-control workplaces shape the minds of those who serve in them.

The study surveyed 3,891 members of the New South Wales Police Force, including both sworn officers and unsworn staff across ranks and roles, with participants averaging 44.3 years of age and 16.7 years of service. The theoretical engine of the work is the marriage of Self-Determination Theory, which holds that human flourishing depends on the satisfaction of three basic psychological needs — autonomy, competence and relatedness — with the Job Demands–Resources model, which describes how workplace demands deplete energy while resources such as support and discretion build engagement and resilience. Previous research had suggested the two frameworks were compatible, but their integration within rigidly hierarchical organisations had remained underdeveloped. The new study fills that gap at unprecedented scale, modelling six demands and resources, six psychological need variables and six outcomes spanning organisational commitment, intention to leave, well-being, life satisfaction, post-traumatic stress symptoms and mental health.

A central methodological innovation lies in how the researchers measured the psychological needs. They used the Basic Psychological Need Satisfaction and Frustration Scale, adapted so that every item referred explicitly to the work context — for example, “At work, I feel a sense of choice and freedom in the things I undertake.” The instrument distinguishes not only three needs but two valences for each: satisfaction and frustration. Contemporary Self-Determination Theory treats these as partially independent processes rather than simple opposites. Need satisfaction feeds adaptive functioning, engagement and well-being, whereas need frustration — the active thwarting of volition, effectiveness or belonging — predicts ill-being, dysregulation and burnout through a separate pathway. To capture this dual-process architecture, the team employed a sophisticated analytic strategy: set exploratory structural equation modelling with target rotation, estimated with robust maximum likelihood in Mplus, combined with an ANOVA-like decomposition of the six need scales into five orthogonal contrasts.

That contrast decomposition deserves technical attention because it resolves a long-standing measurement problem. When the six need subscales are used as separate predictors, their high intercorrelations produce multicollinearity that blurs whether effects reflect the specific need involved, the valence of the experience, or an interaction between the two. The researchers re-expressed the six scales as mutually orthogonal, zero-sum contrasts: one capturing overall valence, two capturing content differences (autonomy versus competence and relatedness combined, and competence versus relatedness), and two capturing valence-by-content interactions. The 18-factor measurement model fit the data remarkably well, with a root mean square error of approximation of just 0.020, comparative fit index of 0.942 and reliability coefficients averaging 0.91. All five contrasts proved statistically significant predictors across all six outcomes, demonstrating that valence, content and their interaction each carry distinct, theoretically meaningful information.

The substantive findings paint a coherent dual-pathway picture. On the resource side, organisational support emerged as the standout job-related resource, strongly associated with need satisfaction — its correlation with autonomy satisfaction reached 0.56 — and, through that satisfaction, with job commitment, well-being and life satisfaction. Personal resources mattered too: problem-solving coping, social-support seeking and mindfulness contributed to adaptive outcomes, with problem-solving showing a notable link to competence satisfaction. On the demand side, workplace conflict functioned as a classic hindrance-type demand, correlating 0.55 with relatedness frustration and predicting need frustration, PTSD symptoms and poorer mental health. Avoidant coping, treated as a personal demand, was among the most salient predictors of adverse outcomes. Predictors explained on average more than 40 percent of the variance in the six psychological needs, underscoring how powerfully workplace conditions shape basic psychological experience.

Mediation analyses then tested the theoretical heart of the integration: whether psychological needs transmit the effects of demands and resources to outcomes. They largely do. Need satisfaction and frustration mediated substantial portions of the associations, with mediation strongest for job-related predictors of job-related outcomes — organisational support flowing through need satisfaction to commitment, workplace conflict flowing through need frustration to intention to leave — and weaker but still meaningful for personal outcomes such as life satisfaction and mental health. The valence-matching predicted by the dual-process model held throughout: need satisfaction most strongly predicted beneficial functioning, need frustration most strongly predicted impaired functioning, and frustration often exerted disproportionately strong effects on adverse outcomes. Autonomy satisfaction showed a particularly marked association with job commitment, while competence frustration loomed large for PTSD symptoms and impaired mental health.

One finding carries particular weight for the world’s police services and other hierarchical institutions: autonomy satisfaction was achievable, and mattered enormously, even within a command-and-control organisation often assumed to be structurally hostile to autonomy. The researchers report that during survey development a senior officer objected, “We can’t let officers decide for themselves, or it would be chaos.” The study’s authors argue this reflects a fundamental misconception. Autonomy in Self-Determination Theory does not mean unrestricted freedom or the absence of rules; it means experiencing volition and ownership within clearly defined parameters. Strict rules governing firearm discharge, for example, do not contradict autonomy, because officers still exercise judgment in how to implement them. Autonomy also encompasses feeling heard — having input, receiving rationales for decisions, and exercising constructive discretion, even when final authority rests with leadership.

The practical implications follow directly. The study recommends that leaders create structured opportunities for officers to contribute insights into procedures, scheduling and operational improvements; explain the rationale behind policies rather than defaulting to command; encourage discretion within clearly articulated operational boundaries; train supervisors in autonomy-supportive behaviours such as acknowledging perspectives and avoiding micromanagement; and invest in competence-building training that lets autonomy and effectiveness reinforce one another. Notably, the demand-by-resource interactions central to classic Job Demands–Resources theory — the supposed buffering effects whereby resources weaken the impact of demands — contributed almost nothing in this sample, explaining on average less than half a percent of additional variance. The authors interpret this as consistent with recent reviews showing that buffering effects in Job Demands–Resources research are typically small and unstable, whereas the additive, need-mediated pathways are robust and far more consequential.

The findings also resonate with Conservation of Resources theory, suggesting that organisational and personal resources initiate gain spirals — more support leads to more need satisfaction, which fuels well-being and commitment — while hindrance demands and avoidant coping drive loss spirals in which need frustration and deteriorating mental health reinforce each other. Yet the authors are candid about limitations. The design was cross-sectional, so the results represent associations consistent with theory rather than demonstrated causal mediation; distressed personnel may perceive their workplaces more negatively, a “gloomy perception” bias that could inflate some associations; response rates were unavailable, so representativeness cannot be confirmed; life satisfaction was assessed with a single item; and the sample derives from a well-resourced Western policing system whose generalisability to other contexts is uncertain.

Even with those caveats, the scale and analytic rigour of the study make it a landmark for occupational psychology in high-stress, command-and-control settings. Its implications extend well beyond policing to emergency services, healthcare and the military — anywhere operational discipline coexists with human beings who need to feel effective, connected and authorship over their own actions. The message to leadership is quietly radical: the choice is not between discipline and well-being. Structure, when delivered with rationale and respect for volition, can support the very needs that protect officers from the chronic stress that, as Sergeant Klein foresaw, is far more likely than a bullet to strike them down.