A review of new science, in plain language

Science  ·  9 July 2026  ·  From issue 06

A statistical test for the benefits of personalizing interventions

A new statistical test can show if custom treatments beat one-size-fits-all options.

Zhaoqi Li, Emma Brunskill.

A new tool checks if giving people different, custom treatments works better than giving everyone the same single best option. It uses past data to make this check. The test keeps a low rate of false alarms. It was tried on data from job training, depression care, schools, and recommendation systems. In these tests, it worked well and beat other methods.

Integrity screen: passed (3 checks) Checked 15 July 2026. Read the source

The plain-language reading of this paper is in the full issue, for subscribers.

Read the full issue

Five vetted papers in plain language, free, every two days

Five papers free every issue, about 15 emails a month. No spam. Leave in one click.

Also in issue 06 9 papers
More from Science