Hacker News new | past | comments | ask | show | jobs | submit login
Run Chart Rules for Interpretation (2016) [pdf] (scot.nhs.uk)
15 points by Tomte on Feb 10, 2020 | hide | past | favorite | 4 comments



From the very first paragraph,

> For example, for an event with two possible outcomes, where each outcome is likely to happen 50% of the time, the probability of the same outcome occurring six times in a row is less than 3 in 1000

This is wrong. The probability of, e.g. getting heads 6 times in a row is 1/2^6 which is 1/64. The probability of getting 6 heads in a row or six tails in a row is 1/2^5, which is 1/32.


Strange. The document claims it's paraphrasing two sources, one of which is linked[0] in the document itself.

The linked source does contain the diagram your quote is associated with, but not the (agreed, questionable) commentary about probabilities.

[0] - http://www.knowledge.scot.nhs.uk/media/CLT/ResourceUploads/1...


You're right.

The mentioned likelihood would be for 8.4 times in a row:

3:1000 → 1:333⅓

P = 1 / (1 + 333⅓)

n = |log(1/334⅓)|

n ≈ 8.3851


For a fuller discussion of statistical process control charts in the NHS (an English NHS example this time) you might be interested in "Making Data Count". https://improvement.nhs.uk/resources/making-data-count/

Especially the Plot the Dots booklet: https://improvement.nhs.uk/documents/2748/NHS_MAKING_DATA_CO...




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: