Skip to content
Plinth

Glossary

Left-censoring

The first year of any dataset makes existing relationships look new.

A funding relationship that began before the earliest year on record appears, in that first year, to be starting. There is no way to distinguish a genuinely new grantee from a long-standing one whose earlier grants sit outside the window.

This is why the opening year of a trend is read with care and why a first-year intake wave is rarely meaningful. The effect only touches the earliest year; later years have prior history to compare against.

Related terms

Questions and answers

What is a left-censoring?
The first year of any dataset makes existing relationships look new. A funding relationship that began before the earliest year on record appears, in that first year, to be starting. There is no way to distinguish a genuinely new grantee from a long-standing one whose earlier grants sit outside the window.

More from the funding graph