Social and Critical Psychology

Constructing a Category

Six messy fictional accounts and five decisions. By the end there is a named condition with a boundary, a questionnaire, a prevalence figure, a training course and a place in somebody's spreadsheet - and nothing new has been discovered.

Simulated - fictional accounts, a fictional survey population and an invented category

Learning objective

By the end you should be able to say where a classification acquires the appearance of a natural object, and why “socially constructed” does not mean “made up”.

About 25 minutes. Nothing you do here is saved or sent anywhere.

Construction is not fabrication

The six people describe something. They are tired, they dread opening their messages, they stay up late for an hour of quiet. Nothing in this activity suggests they are inventing it, and nothing here is a criticism of anyone who finds a category useful for naming their own experience. The question is a different one: what happens to a description once it is given a name, a threshold, an instrument and an institutional user - and what stops being asked as a result.

How to use this

Read the six accounts.
Commit to an answer to the question below.
Work through the five stages in order. Each adds one result to the panel tracking the category you are building.
At the end, go back and change one decision - the threshold is the most instructive - and watch what moves.

All six are invented for teaching. No real respondent, dataset, survey or published category is reproduced.

This part needs JavaScript

The laboratory recomputes coverage, prevalence and institutional uptake in your browser as you make each decision, so it cannot run with JavaScript switched off. The accounts, the debrief and the limitations are readable without it.

The team will turn these accounts into a category and then estimate how many people have it. What will most determine that figure? How common the experience actually is in the population How well the six accounts represent everyone else Where the team decides to draw the boundary How good the questionnaire turns out to be

Need a hand?

Key terms

Reification: Treating the product of a classification as a thing that was there all along waiting to be found.
Boundary: The rule deciding who is inside the category: how many features, for how long, with how much interference.
Prevalence: How many people the category contains. It depends on where the boundary was put as much as on the people.

Stage 1. Which features recur?

Tick the features you think define what these six people are describing. Choose at least three. The table beside you shows how often each one appears here, and how often it appears in a comparison set of accounts from people who said nothing was difficult to name.

Stage 2. What will you call it?

Nothing about the accounts changes here. Everything about what follows does.

Stage 3. Where is the boundary?

Applied to 400 simulated respondents. The model is described in the notes at the foot of this panel.

The other two conditions in the rule

Stage 4. Which items go in?

Choose five items for a short screening questionnaire. The result shows which of the original six accounts your instrument would still find.

Stage 5. Things that happen next

You no longer control the category. You control only what you do when somebody else picks it up.

Three more decisions land on your desk

The category so far

What your features pick out

Each selected feature, the number of the six accounts it appears in, and the number of six comparison accounts it appears in
Feature	In the 6	In 6 comparisons	Distinguishing?

How many people have it

Number and percentage of the 400 simulated respondents in each band
Band	People	Share

Who your instrument still finds

Each of the six original accounts and whether the questionnaire would detect it
Account	Items it would endorse	Detected?

The six accounts, for reference

What one step of the threshold does

Prevalence at every possible feature threshold, holding the other two requirements as set
Features required	Prevalence

Reification is usually taught as an attitude — treating an abstraction as a thing — and left there. Here it has steps you can point at: selecting features fixed what would count, naming imported a frame and an expert, a threshold turned a continuum into two kinds of people, and an instrument made the category prior to the experience it came from. Nobody was careless and nothing new was discovered.

The material

First, a judgement

The object-status laboratory

Five stages from a description to a thing

The category so far

What your features pick out

What the name imports

How many people have it

Who your instrument still finds

Object status

What happened, and what stopped being asked

Challenge - five sentences about your category

What this demonstrates