this post was submitted on 15 Nov 2023
440 points (93.5% liked)
Technology
59068 readers
3471 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Why do you think the conclusions aren’t fair? Does the sample seem biased? I’m very confused as to what you’re getting at.
Statistically speaking, that’s a perfectly fine sample size (large even). It sounds like it’s fairly representative. You can play around here if you’d like: https://www.qualtrics.com/blog/calculating-sample-size/
Private dealerships tend to be more heavily focused on used car sales in my understanding.
Statistics or not, the fact remains that there were thousands of dealerships not accounted for. Human choices and interactions that weren’t experienced. So to generalize to every dealership or even most dealerships is dismissive of human choice and paints an inaccurate picture.
The correlations they draw from this data are far out of the bounds of what the stats actually say and what they do not say. Their data acquisition is sparsely outlined, this isn’t a research paper that was properly conducted. It’s a rando poll that all these news orgs like to pop up to mislead the public by bending the data to fit their narrative. Par for the course of the Washington post.
So we cannot learn anything from representative sampling? I'm still not following. You just invalidated like 99% of published studies with that argument
In the link you provided, the methods seem clear enough to me?
One glaringly obvious “oddity” was the fact that they did representation based on state by state average, but then lumped each state into a group and then tried to say that was representative of the nation? How does that follow? Should it not be based off the nations as a whole if we’re making those generalizations? Margin of error being 3% of thousands is a lot of dealerships.
All I’m saying is that people need to excercise caution when it comes to statistics. These might be statistically sound measurements but the story they actually tell is much more specific. The data only tells you what you measured, meaning is derived.
For example, remember the chip shortage? Wonder why they all had such a hard time getting those cars in. Wonder how many of those dealerships if they could get EVs would sell them regardless of the opinion of one franchise owner - because business swings at its own pace. If that owner pushed against them they’d just fire him and hire someone else who won’t have a problem with it.
Maybe they don’t sell EVs because the populace legitimately doesn’t want one.
They also mention “translating answers to yes or no” what does that entail? Why was a critical data transformation not explained in detail??
They make no effort to explain any potential conflicts or any errors with their paper whatsoever- this is a boutique poll presented inappropriately and will be misconstrued by the media as always and echoers.
Don’t lump this poll with real data science.
It would be a lot easier for you to just learn the basics of statistics dude
K
Sigh. Your first post started off so intelligently too, then you had to respond.
Will you elaborate so we can discuss where I went wrong or did you just post this as an insult to feel better about yourself (idrk why else you’d come with a fork to people quietly eating their soup)
Right here
Poor wording at best - what exactly in my analysis is wrong regarding the sampling size. They based off state by state, then grouped into regions and then tried to generalize to the whole country, despite not talking to any private dealerships which are a significant populace.
Im being genuine too, nobody has specifically said what was wrong and what the “right interpretation” was.
Not one. You’ve gotta understand where I’m coming from on this aspect id hope. This isn’t sarcasm or anything else - please work with me to correct my understanding.
Why would you account for used car lots in a study about buying a brand new EV from a company dealership?
Because the headline is saying that dealerships won’t sell them period. Which would include used ev vehicles, which is my entire point. The data they collected made little to no effort to distinguish exactly what they were measuring for - new sales by franchised dealerships that were sample sized based off state to state (region to region) as an explanation for a national average/consensus. It doesn’t make sense to me but maybe there’s an aspect about sample sizing I’m not fully grasping.
Believe it or not I too went through high school science which specifically and religiously looked at statistics p error and all that fun stuff. Perhaps an aspect has faded from memory and I’m way off but I have yet to have 1 personal actually explain where I’m missing understanding or just flat out wrong just quips and insults so I’m not so convinced. Where am I misunderstanding?