Invisible Women: Data Bias in a World Designed for Men
A writer examines how missing or poorly separated data about women can shape products, workplaces, public services, and research around male defaults.
Caroline Criado Perez examines situations where information about women is absent, inadequately collected, or folded into assumptions that treat male experience as the general norm.
An editorial distinction separates dataset size from representation. Many observations can still omit relevant experiences; asking who was measured, under which conditions, reveals more than quantity alone about possible fairness.
The book connects this data gap with everyday environments, including transport planning, workplace arrangements, product design, and the organization of paid and unpaid labor.
A default can seem neutral because it is familiar, even when the underlying measurements fit some bodies, schedules, or responsibilities better than they fit others.
Care work becomes important to the analysis because travel patterns and time demands may be poorly represented when planners imagine an uncomplicated journey between home and employment.
Research and safety testing also raise questions about whose characteristics were included and whether results were examined separately before being generalized across a wider population.
Inclusion should broaden inquiry rather than install another universal person. This reading pairs separated data with variation within groups, avoiding assumptions that a shared demographic label means identical needs.
Criado Perez argues that these omissions often reflect inherited habits rather than an explicit intention to exclude, although their consequences can still be serious and unequal.
The examples support asking better questions about evidence, not assuming that every woman has the same needs or that sex explains every difference between people.
Race, disability, age, income, and other circumstances interact with gender, making a single replacement average inadequate as a complete solution to exclusion in design.
Individual statistics and examples belong to specific studies and periods, so current decisions need updated evidence rather than uncritical repetition of memorable figures from the book.
Its central invitation is to notice whose experience a dataset makes visible, and to build institutions that can discover important omissions before they become routine harms.
Race, disability, age, income, and other circumstances interact with gender, making a single replacement average inadequate as a complete solution to exclusion in design.
How should designers identify missing experiences without reducing the people they hope to include to another oversimplified demographic average?