From the parents cited article (aka, 2021 data) states:
> The American Customer Satisfaction Index independently measures customer satisfaction with top-selling smartphone brands available to U.S. consumers.
Smartphone brands is what I’m referring to as “comparing phone hardware”.
That article doesn’t asses satisfaction with the two successful proprietary ecosystems, directly.
By measurement artifact, I mean that the appearance of consistency could actually be artificial statistical noise hiding a confounding variable.
ie, did they actually ask questions that compare satisfaction with App Store vs Play Store or only ask for satisfaction with Galaxy verse Pixel verse iPhone, etc.
That's a lot of mental gymnastics you're doing there.
Why would a question about satisfaction with your smartphone exclude the entire operating system? I mean, maybe that's what they did(?). But there is no evidence for that. At all.
I'm not even going to respond to the wild speculation (again, without evidence) that the results are flawed by a "measurement artifact".
> That's a lot of mental gymnastics you're doing there.
Indeed, scientific reasoning does tend to bend the mind.
> Why …
What is at issue is the scientific validity of drawing conclusions about satisfaction with an ecosystem from data about satisfaction with phone models.
> I’m not even going to respond…
But you did respond with wild accusations, in violation of HN site guidelines.
Since you didn’t take the charitable effort to Google “measurement artifact”, here you go, allow me to provide a Wikipedia article on artifacts: https://en.wikipedia.org/wiki/Artifact_(error)
> an artifact is a spurious finding, such as one based on either a faulty choice of variables or an over-extension of the computed relationship. Such an artifact may be called a statistical artifact.
>>an artifact is a spurious finding, such as one based on either a faulty choice of variables or an over-extension of the computed relationship. Such an artifact may be called a statistical artifact.
I know exactly what was meant, the problem is you've provided absolutely no evidence that it is a measurement artifact(!). You simply stated "it might be". What kind of "scientific reasoning" is that exactly? Please tell.
The onus is on you to provide evidence to back up your claims, or admit that you flippantly dismissed study results simply because they run counter to your point.
That is what is at issue. I made a point and provided supporting evidence. You have not. I made no "accusations" other than that.
>Also, the 2021 data being referenced is comparing phone hardware, not ecosystems.
I didn't see that mentioned anywhere so I'm going to ask for a citation. It appears to me that the study asked for an overall impression.