Path: ...!feeds.phibee-telecom.net!2.eu.feeder.erje.net!feeder.erje.net!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Your Name <YourName@YourISP.com>
Newsgroups: misc.phone.mobile.iphone
Subject: Re: green bubble syndrome
Date: Sun, 13 Oct 2024 10:44:51 +1300
Organization: A noiseless patient Spider
Lines: 34
Message-ID: <veeqki$at9n$1@dont-email.me>
References: <xn0oruv2k1siabt002@reader443.eternal-september.org> <ve6sv0$2q45v$1@dont-email.me> <veasft$3k74p$1@dont-email.me> <veduk6$65bo$2@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Sat, 12 Oct 2024 23:44:52 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="aa12fcf3d668ef3d42689a3ca0a76ce4";
	logging-data="357687"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19IAgG/s96y5lgnK5epJ5YMloIFLu81Wao="
User-Agent: Unison/2.2
Cancel-Lock: sha1:JzD3dsAkvF6llhfhdecGfyjRZ2E=
Bytes: 2592

On 2024-10-12 13:46:48 +0000, Wilf said:
> On 11/10/2024 at 22:14, Jörg Lorenz wrote:
>> On 11.10.24 21:11, Chris wrote:
>>> Wilf <wilf21@is.invalid> wrote:
>>>> On 10/10/2024 at 17:28, Jolly Roger wrote:
>>>>> Apple sold 2.5 BILLION iPhones (as of 2023, so not counting 2024), and
>>>>> you are trying to tell us that a survey of 1000 people is significant?
>>>>> Quick question: How many times do you think 1000 goes into 2.5 billion?
>>>> 
>>>> If the sample is chosen properly (and that's the critical part), results
>>>> from a small but representative sample of the whole population can be
>>>> statistically significant.  So just because someone has no background in
>>>> statistics is a not a reason to necessarily doubt the premise.
>>> 
>>> Correct.
>> 
>> No. It is not correct by any means.
>> I still lack the proof that the sample is relevant.
> 
> That's my point.  We have to be persuaded that the sample was properly 
> chosen.  Beyond that, the relatively small sample size does not in 
> itself invalidate the significance of the results.

The results are only valid for the 1000 people they bothered to survey. 
Claiming anything above that is pure nonsense and at best simply a very 
rough guesstimate.

Such surveys should always specifically say something like "40% of the 
people surveyed" ... *NOT* *EVER* simply "40% of people", because 
that's where the novice general public then confuses it with being 
reality for everyone, when it is not.