# @theonecoleman on Instagram

- **Type:** Image
- **Original URL:** https://www.instagram.com/p/DeModl8IKj0
- **Gondola URL:** https://gondola.cc/posts/71541722-theonecoleman-instagram
- **Thumbnail:** https://img.gondola.cc/tr:w-,h-,fo-auto/postThumbnails/a02444c683.jpg
- **Posted:** 2026-10-07T15:01:09.000+00:00
- **Account Owner:** Theone Coleman (@theonecoleman) — https://gondola.cc/theonecoleman

## Caption

90% said they improved. 90% of who?

It sounds impressive. Here’s a made-up example of what could sit underneath it.

One hundred people start. Sixty answer the follow-up. Fifty-four say they improved. That’s 90% of the people who replied. Measured against everyone who started, it’s 54% reporting improvement.

And the other forty? We don’t know. They’re not automatically failures. But you can’t quietly count them as successes either, or forty people just disappear behind one good-looking number.

That 90% can be accurate. It just can’t stand on its own.

So put the three counts together: who started, who answered, who said they improved. Say what “improved” means. Then look for the reasons people couldn’t, or didn’t, answer. That may change what you need to fix next. The calculation can be right while the impression is wrong.

Worked example: the numbers are invented to show the arithmetic.

On why missing replies matter: HM Treasury Magenta Book, section 4.3.2 (survey data and non-response)

Want to stay updated on what’s coming next? DM me your email.

#WhoBenefits #EvidenceMatters #SocialImpact

## Stats

- **Views:** 0
- **Likes:** 1
- **Shares:** 0
- **Comments:** 0

## Tags

whobenefits, socialimpact, evidencematters

---
Copyright (c) Gondola