#yearofeading

‘Invisible Women’ explores how missing or misused data leads to a world that fails women, often with life threatening consequences. Although this book is several years old, it still serves a useful role by identifying areas where data on women has historically been overlooked.

My first book for 2026 is ‘Invisible Women’ by Caroline Criado Perez. As someone with a keen interest in the use of data for decision making,  I’m always interested in books that examine where the weak points are in the data life cycle. 

The default male and missing women

‘Invisible Women’ highlights how research and data collection often assume a ‘default male’, which can lead to decisions being based only on what is optimal for men. This is, as shown through numerous examples in the book, rarely what is optimal for women. 

Sometimes missing data arises when, (typically male) researchers cite concerns about the cost or complexity of conducting trials and experiments that include women. In others, the lack of women in government or industry means problems impacting women are not even recognized, and hence no data exists on what may be a long standing issue.

In other cases, there appears no clear explanation for why data on gender wasn't captured or reported on. As someone who has worked in industries with rich data, including data on gender, I was surprised to discover that many sectors didn't collect or publish analysis with a gender split.

Perez also highlights examples of where even when data on women was available, this data was not used properly or was outright ignored.

The assumption of the ‘default male’ perspective means that without a conscious effort to include more women in public, political and corporate decision making, existing blindspots will remain.

Lessons for researchers

The two simplest lessons for researchers that emerge from this book are: 1) don’t assume the ‘default male’ or the gender neutral human will do. Women need to be part of the data set.  2) Once you’ve collected data on women, this should be included as one of the key data splits. Even if there is no difference in results, this should at least be mentioned. Do not simply lump all results into one category of ‘people’.

Sometimes less is more when persuading with data or its absence

For a book about the lack of data on women, ‘Invisible Women’ is at times a firehose  of data points and citations. As someone who values data, I appreciate the author’s diligence in bringing together so many examples to back up each of the thematic areas the book examines. 

The persuasiveness of the book might potentially have been improved on though by doing a deeper dive on some of the case studies. Examples of highly impactful policy or research proposals being passed over by grants bodies might have benefited from an in text explanation of what alternate proposals received funding instead.

A worthwhile read for anyone interested in improving data driven decision making