this post was submitted on 03 Apr 2024
145 points (90.1% liked)

Technology

1377 readers
189 users here now

Which posts fit here?

Anything that is at least tangentially connected to the technology, social media platforms, informational technologies and tech policy.


Rules

1. English onlyTitle and associated content has to be in English.
2. Use original linkPost URL should be the original link to the article (even if paywalled) and archived copies left in the body. It allows avoiding duplicate posts when cross-posting.
3. Respectful communicationAll communication has to be respectful of differing opinions, viewpoints, and experiences.
4. InclusivityEveryone is welcome here regardless of age, body size, visible or invisible disability, ethnicity, sex characteristics, gender identity and expression, education, socio-economic status, nationality, personal appearance, race, caste, color, religion, or sexual identity and orientation.
5. Ad hominem attacksAny kind of personal attacks are expressly forbidden. If you can't argue your position without attacking a person's character, you already lost the argument.
6. Off-topic tangentsStay on topic. Keep it relevant.
7. Instance rules may applyIf something is not covered by community rules, but are against lemmy.zip instance rules, they will be enforced.


Companion communities

[email protected]
[email protected]


Icon attribution | Banner attribution

founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] [email protected] 140 points 7 months ago (3 children)

For those unfamiliar, each month, Valve randomly picks a subset of Steam users and offers them anonymous participation in the survey. ... According to Valve, in March 2024, Windows 11 went down from 41.96% to 41.61%. On the other hand, Windows 10 gained what Windows 11 lost—its market share increased from 54.19% to 54.40%.

Eesh, I dunno. Those are quite small changes and could just be a result of statistical noise, especially given the random sampling.

Not exactly headline worthy, IMO, but it must be a slow news day.

[–] [email protected] 11 points 7 months ago (2 children)

It's a small change, but the trend should be decisively in the opposite direction. New Windows OSs are almost always growing in the market relative to old Windows OSs. This means Windows 11 is not popular and Windows users are avoiding it.

[–] [email protected] 12 points 7 months ago

I'm willing to bet it's just the tpm requirement holding off growth. People aren't buying hardware just to install win 11.

[–] [email protected] 6 points 7 months ago

This small of a change is probably within a margin of error.

[–] [email protected] 4 points 7 months ago

Well, the random sampling makes it less likely it's invalid. Random sampling is pretty much the standard for large groups. As for significance, that's something I'd have to read more to know whether it's high enough to truly matter. There are very well developed rules for this stuff in statistics.

[–] [email protected] -2 points 7 months ago* (last edited 7 months ago) (1 children)

Do you know what the sample size for it is? Probably in the millions i would expect. If thats the case then i dont think 0.3% is just noise, but it also doesnt really show anything of value.

[–] [email protected] 18 points 7 months ago (1 children)

I can't find anything on the sample size. It's not mentioned on the official page, and all the other results are armchair statisticians arguing about it in various forums. I'm guessing they want to keep that data close to their chest.

However, Steam's charts page shows a peak of 34M players online at once over the last few days. A few different sites suggest Daily Average Users are around 60M. Let's call it an even 50M for the sake of argument.

What would a decent sample size be without generating overwhelming amounts of data? Say, 10%? So that's surveying 5M users.

0.3% of 5M is just 15,000 users. What if the survey just happened to pick 15k fewer Windows 11 users this time? Is that so unbelievable?

[–] [email protected] 1 points 7 months ago (1 children)

Yeah that is insanely unbelievable. Maybe a tenth of that is barely realistic but 15k of sample error would mean that valve is completely shit at selecting a good sample.

[–] jacksilver 3 points 7 months ago (1 children)

Even if their sample size is large, you're assuming that people who are willing to respond to the survey are evenly distributed across different OS. I wouldn't be surprised if their is a slant to this data just because it's based on an opt in survey.

[–] [email protected] 1 points 7 months ago

That is a good point and might be true, but statistics people deal with that kind of bias all the time and can also adjusted for it. But yeah who knows what kind of people are responsible for these surveys at valve.