Organisational PsychologyWhat It Measures, and What It Leaves Out

Between 1928 and 1932, researchers at the Western Electric Hawthorne Works outside Chicago conducted more than twenty-one thousand interviews with employees. The interviews began at around half an hour and stretched, over the years, to ninety minutes and sometimes two hours, because the researchers found that workers allowed to talk at length felt better afterwards. The programme’s own archive describes the process as one meant to provide an emotional release.

One of the founding scenes of organisational psychology, then, is a company listening to its workers with enormous patience and at considerable expense, so that they would feel better about work whose conditions largely stayed as they were. The scene contains most of what the discipline would become: genuine curiosity about people at work, serious method, real benefits to the people studied, and a client whose interests set the terms of the inquiry.

Organisational psychology has grown into a rigorous applied science since then, and much of working life is better for it. What follows concerns something narrower: the things the discipline, by the shape of its methods and the position of its practitioners, finds hardest to see. Those blind spots follow from where the discipline usually stands: inside the organisation it studies, hired by it, and measuring one person at a time. The view from that position is detailed and honest, and it has edges.

What Organisational Psychology Studies, and Who It Studies For

Organisational psychology, also called industrial-organisational or I-O psychology in the United States and occupational psychology in the UK, is the scientific study of people at work. It covers selection and assessment, training, motivation, leadership, teams, performance, wellbeing and culture. Its professional home in the United States is the Society for Industrial and Organizational Psychology.

The discipline was born answering practical questions. Hugo Münsterberg’s Psychology and Industrial Efficiency, published in English in 1913, proposed applying the new experimental psychology to choosing and managing workers. During the First World War, American psychologists built the Army Alpha and Beta tests to sort hundreds of thousands of recruits. Between the wars, the Hawthorne studies turned attention from the physical conditions of work to its social and emotional life and helped launch the human relations movement. Every one of these moments began with an organisation that needed an answer: whom to hire, where to place them, how to get more from them, why they were leaving.

That pattern has held. The organisation is usually the client, and the client’s questions become the discipline’s research agenda. Nobody should find this sinister, since applied sciences follow their funding. But it has consequences. The unit of analysis tends to be the individual, because individuals are what organisations select, train, assess and replace. The outcomes that matter tend to be the ones the organisation already values, such as performance and retention. And the findings travel upward, to the people who commissioned them, before they travel anywhere else. The workers who supplied the data are usually the last to see what was concluded from it, if they see it at all.

Selection shows how quietly this operates. A psychometric assessment decides who gets in, which makes it one of the most consequential belonging decisions an organisation makes. It is presented as measurement, and much of it is careful measurement. It is also a definition, written in advance, of the kind of person the organisation wants to become part of it. Whoever writes the definition shapes the membership, and the candidates meet the instrument long before they meet anyone who could explain it.

The discipline did grow a second lineage that looked past the individual, and its fate is instructive. In 1951 Eric Trist and Ken Bamforth of the Tavistock Institute published a study of coal mining in the north of England. Mechanised longwall mining had broken up the small, self-regulating teams that had worked the older coal faces together, replacing them with specialised shifts each doing one fragment of the cycle. Absenteeism and conflict rose, and output suffered. The problem sat in the fit between the technology and the social organisation of the work rather than in the miners’ attitudes. Their sociotechnical systems thinking influenced a generation of work design, yet the mainstream of the discipline moved in the other direction, towards the individual and the measurable trait. Instruments are easier to sell and scale than an inquiry into how a whole system has arranged itself.

The same preference now runs at machine speed. People analytics platforms collect sentiment from pulse surveys and risk scores for who might leave. The measurement habit the discipline built by hand is being extended by software into something close to continuous observation, and the individual remains the unit being scored. The Algorithmic Shadow follows where that leads. The older question comes along unchanged: who designed the measures, and what the people being measured know about them.

The Hawthorne Studies and the Birth of Listening as Management

The Hawthorne studies began with light. From 1924 to 1927, engineers varied the illumination in three departments to see whether better lighting raised output. The records held at Harvard Business School’s Baker Library report that the tests found no significant correlation between light and productivity. Something else was going on.

The researchers moved to a relay assembly test room, where six young women assembled telephone relays under close observation for five years while rest periods, hours and incentives were varied around them. Productivity rose, and the explanation that became famous, that people respond to being watched and cared about, entered the textbooks as the Hawthorne effect. When the economists Steven Levitt and John List went back to the original illumination data in 2011, they found far less there than the famous story claimed. The legend outlived the evidence because it was useful. It told managers that attention could raise output, which is a cheaper finding than most. The sociologist Alex Carey pressed the point in 1967, arguing in the American Sociological Review that the human relations conclusions drawn from Hawthorne rested far less on the evidence than on the convictions its interpreters brought to it. Elton Mayo, who led the Harvard side of the work, saw industrial conflict largely as a failure of social cohesion and workers’ resistance as feeling to be understood and soothed. That reading suited management well, and it shaped how the discipline would understand the emotional life of work for decades.

Read from another angle, Hawthorne teaches a less comfortable lesson. It discovered that the social and emotional life of workers mattered to production, and in the same movement it turned that life into something to manage. The interview programme is the clearest case. Workers spoke about their supervisors, their families, their money worries and their fears, often for hours, and the stated value of all that speech lay in what it did for their mood at work. Listening became a management technique. The feelings were real, and they were being put to work.

A Western Electric supervisor, Donald Chipman, wrote down the other lesson in 1931: figures don’t lie, but we have shown that we can take a set of figures and prove anything we want to. The discipline was born carrying its own warning label. Every generation since has had to decide how seriously to read it, usually under pressure from someone who wanted the figures to prove something in particular.

What Organisational Psychology Leaves Out

The easiest way to see what a discipline struggles with is to look at the work sitting on its edge.

In the late 1950s the Tavistock Institute was asked to help a London teaching hospital with a staffing problem: student nurses were leaving. Isabel Menzies published her findings in Human Relations in 1960, and they remain some of the most unsettling pages in organisational research. The hospital, she found, had organised nursing in ways that protected nurses from the anxiety at the heart of the job, which was sustained closeness to suffering, bodily intimacy and death. Care was split into tasks allocated by list rather than by patient, so that no nurse looked after a whole person. Decisions were checked and counter-checked until responsibility dissolved. Students were moved between wards before attachments could form.

Inside the hospital these arrangements were simply how nursing was done, and nobody experienced them as defences. And they failed on their own terms. The anxiety stayed, while the work lost the meaning that had drawn people into it in the first place: the chance to care for someone. The students who left were, in a sense, the ones who could still see what had been removed.

Imagine the same hospital running a modern engagement survey. It would likely have found low scores on recognition and workload, and recommended action plans for each. It could not have found what Menzies found, because her finding lived in the structure of the work itself, in a system organised, below anyone’s awareness, to keep feeling at a distance, and no individual’s answers contained it. Survey instruments average individual responses and report the result. They are poorly built to notice an organisation doing something, as a whole, that none of its members would choose.

This is where the omissions begin to overlap. Anxiety of that kind is organisational, and organisations tend to manage it by converting it into something more tractable, usually an individual deficit. Stress becomes a resilience gap. Conflict becomes a personality clash to be decoded with a type indicator. Each conversion moves the difficulty away from the structure and into a person, where training or an app can reach it. The discipline supplies many of the tools for these conversions, often with the best intentions, and they are sometimes exactly what is needed. They are also exactly what an organisation defending itself against a harder truth would want.

Power runs through the same territory, though the discipline tends to study it as a variable rather than as the medium it works in. Leadership style and decision rights are well researched. Far less examined is the fact that every instrument lands inside a relationship of power. An engagement survey is designed and read by the organisation, and the people completing it know it. What they write is shaped by what is safe to write. Anonymity offers less protection than the covering email promises. In a team of seven, a comment about a restructure, written in a recognisable turn of phrase, identifies its author to anyone who knows the team, and the people completing the survey know that too. So the free-text box fills with careful, generalised language, and the careful language is read back as a finding about how people feel. The results come back as a portrait of employee sentiment, when a good part of what they portray is the organisation’s own capacity to be told the truth.

And meaning sits just past the discipline’s instruments. Psychology measures states well: satisfaction, stress, commitment, intention to leave. It measures less well the stories people tell about what their work is for and whether it still holds together. Menzies’s nurses might well have reported adequate satisfaction. What they had lost was harder to put on a five-point scale: the sense that the job was the job they had chosen. An organisation can hold every measured state steady while that sense drains away, and the first sign it gives is usually a resignation letter that surprises everyone except the colleagues who saw it coming.

Organisational Psychology Through The Fractured Lens

Read this way, organisational psychology is a participant in the systems it studies, which is a harder position to occupy than that of neutral observer and a more interesting one to examine.

Practitioners inside organisations live this position every day. The in-house organisational psychologist or development specialist often sees the system more clearly than anyone, because people tell them things they would not tell a line manager. They are also employed by the system they see. Their findings pass through the hands of the people whose decisions the findings implicate, and by the time a diagnosis reaches the leadership deck it has usually been softened into something actionable, such as a capability gap or a need for training. The softening is the price of being allowed to keep working on the problem, and many good practitioners pay it knowingly and without cynicism.

Measurement changes what it measures. Once engagement is surveyed, it becomes a number. Once it is a number, it becomes a target for managers, who are coached to raise it. Once it is a target, the survey season becomes a small performance, with teams reminded how much their participation matters and results quietly compared across departments. The instrument designed to hear people ends up teaching them what the organisation wants to hear. The Illusion of Psychological Safety traces the same drift in the discipline’s most successful recent export. Amy Edmondson described psychological safety as a shared belief within a team about whether it is safe to take interpersonal risks. Organisations frequently implement it as an individual competence, a behaviour to be trained and a score to be tracked, so that a property of the group’s climate becomes one more thing each employee can be assessed against.

The discipline’s relationship with uncertainty deserves the same attention. Much of its authority rests on measurement, and measurement produces confidence. A validated scale and a benchmark against the sector both feel like knowledge, and some of them are. Psychology as a whole has spent recent years discovering how many celebrated findings fail to replicate, and organisational applications tend to run a few steps behind the research that should temper them. Chipman’s warning from 1931 has never been more relevant: the figures can prove what they are asked to prove, and it matters who is doing the asking.

The Fractured Lens reads these patterns as forces rather than failures. Dynamics decides who commissions the study and who reads the results. Unknowing shows up in the confidence that measurement lends to decisions made in genuine uncertainty. Meaning is what the instruments reach for and rarely hold. Belonging is the quiet calculation every respondent makes before typing a word into the comment box. It is why the Fractured Lens AI Transition Diagnostic scores the gap between paired statements, such as a claim that a change is collaborative set beside a sense that disagreeing would be noted, rather than averaging answers into a mood. The contradictions are where the organisation’s defences show, and where its people are telling the truth in the only form available to them.

The Hawthorne interviewers listened for thousands of hours, and the transcripts of what workers said were kept under restriction for some seventy years before researchers were allowed to read them. Tens of thousands of hours of people describing their working lives, gathered so that they would feel better, and then sealed. It is worth asking what an organisation is listening for when it asks its people how they feel, and what it would do with an answer it could not manage.