Jump to content
talkfootball365
  • Welcome to talkfootball365!

    The better place to talk football.

Coronavirus (COVID-19) Outbreak


football forums

Recommended Posts

Thankfully it's not that much of an emergency in the UK and it's really only getting me in the mood to watch post apocalyptic movies.

Will you join one of the mad gangs of men that drive around doing crazy shit because they can or will you still try to live in a civilised manner with other people?

Link to comment
Share on other sites

4 hours ago, Cicero said:

Of a sample size of 787 beer drinkers. 

CNN - 38% of Americans 

xD

For the record, my post is assuming the survey was just a "yes/no" question but...

Yeah that's obviously not a big enough sample size to be able to claim it "speaks for all Americans" with any high confidence level (you'd need a sample size of 38,412 for a survey with a 95% confidence level & .5 confidence interval). But that's a ridiculous amount of people and I've never ever heard of any kind of survey polling that tests that many people. Granted, typically these surveys break people down more into particular groups.

But with a 95% confidence level & a confidence interval of 3.5 (the higher the confidence interval, the less certainty there is about the population size) you'd only need a sample size of 784 for the US proximation (which this goes beyond, so this survey better confidence interval than 3.5).

So small sample sizes don't necessarily mean the survey doesn't show anything useful. Otherwise surveys would probably never be conducted because they'd be prohibitively expensive.

Link to comment
Share on other sites

44 minutes ago, Dr. Gonzo said:

For the record, my post is assuming the survey was just a "yes/no" question but...

Yeah that's obviously not a big enough sample size to be able to claim it "speaks for all Americans" with any high confidence level (you'd need a sample size of 38,412 for a survey with a 95% confidence level & .5 confidence interval). But that's a ridiculous amount of people and I've never ever heard of any kind of survey polling that tests that many people. Granted, typically these surveys break people down more into particular groups.

But with a 95% confidence level & a confidence interval of 3.5 (the higher the confidence interval, the less certainty there is about the population size) you'd only need a sample size of 784 for the US proximation (which this goes beyond, so this survey better confidence interval than 3.5).

So small sample sizes don't necessarily mean the survey doesn't show anything useful. Otherwise surveys would probably never be conducted because they'd be prohibitively expensive.

Nah I never take these surveys seriously, especially when they try and blatantly mislead. Surveying ‘beer drinkers’ and pulling it off as Americans. What about those that don’t drink? Or those that don’t drink beer?  Sample size also shows no characteristics regarding age, gender, IQ, etc. Just ‘beer drinkers’ and try to poorly represent it as 38% of all Americans. 

Edited by Cicero
Link to comment
Share on other sites

22 minutes ago, Cicero said:

Nah I never take these surveys seriously, especially when they try and blatantly mislead. Sample size shows no characteristics regarding age, gender, IQ, etc. Just ‘beer drinkers’ and try to poorly represent it as 38% of all Americans. Poor attempt. 

I don't think there's anything misleading about it, other than the fact they don't mention the survey's confidence level, confidence intervals, or standard deviation. But I suspect most people aren't going to give a shit about statistics terms, let alone understand them.

Here they took a sample of "beer drinking Americans" - which is the target demographic for Corona beer, as that article indicates that Corona is heavily dependent on the US market. There's a lot of different characteristics that are encompassed in Americans who drink beer - so there's no reason to really ask the respondents their age, gender, IQ, etc... because they don't really care about these sub-demographics. They're looking to take a sample of beer drinking Americans, it's done by a PR firm that's probably been hired by Corona to ask this question to see if the virus has anything to do with their drop-off in sales.

I don't think it's a poor attempt, it's just how you conduct a survey. Obviously a higher sample size would make things better. They probably should mention the confidence levels and all of that stuff that lets you really interpret the data from these surveys in the news - and more people should probably learn about what these things are and what they mean.

Link to comment
Share on other sites

14 minutes ago, Dr. Gonzo said:

I don't think there's anything misleading about it, other than the fact they don't mention the survey's confidence level, confidence intervals, or standard deviation. But I suspect most people aren't going to give a shit about statistics terms, let alone understand them.

Here they took a sample of "beer drinking Americans" - which is the target demographic for Corona beer, as that article indicates that Corona is heavily dependent on the US market. There's a lot of different characteristics that are encompassed in Americans who drink beer - so there's no reason to really ask the respondents their age, gender, IQ, etc... because they don't really care about these sub-demographics. They're looking to take a sample of beer drinking Americans, it's done by a PR firm that's probably been hired by Corona to ask this question to see if the virus has anything to do with their drop-off in sales.

I don't think it's a poor attempt, it's just how you conduct a survey. Obviously a higher sample size would make things better. They probably should mention the confidence levels and all of that stuff that lets you really interpret the data from these surveys in the news - and more people should probably learn about what these things are and what they mean.

They surveyed beer drinkers and beer drinkers only, and in classic CNN fashion they misled the audience by representing it as all Americans. 

Again, a poor attempt. 

Edited by Cicero
Link to comment
Share on other sites

2 minutes ago, Cicero said:

They surveyed beer drinkers and beer drinkers only, and in classic CNN fashion they misled the audience by representing it as all Americans. 

Again, a poor attempt. 

Around 2/3 of Americans drink, 41% of them say they primarily drink beer; a sample size of only beer drinkers doesn't mean that it doesn't reflect a sample size of all Americans. It might indicate flawed sampling, but it can still be used to say "38% of Americans. But there's flaws in literally every sort of sampling like this.

It's more inaccurate to dismiss it as just false statistics you can't take seriously. Considering some of the crazy shit that so many people out there believe, it's not that outlandish that over 1/3 of Americans think coronavirus is caused by a Mexican beer.

Link to comment
Share on other sites

  • Moderator
49 minutes ago, nudge said:

Same here. Bought a crate last week; it's shitty as beer but goes great with spicy food hehe.

And it's a great "dad joke" to take a picture of it and telling your friends "huheuheu, I've got Corona." 

  • Haha 2
Link to comment
Share on other sites

1 minute ago, Dr. Gonzo said:

Around 2/3 of Americans drink, 41% of them say they primarily drink beer; a sample size of only beer drinkers doesn't mean that it doesn't reflect a sample size of all Americans. It might indicate flawed sampling, but it can still be used to say "38% of Americans. But there's flaws in literally every sort of sampling like this.

It's more inaccurate to dismiss it as just false statistics you can't take seriously. Considering some of the crazy shit that so many people out there believe, it's not that outlandish that over 1/3 of Americans think coronavirus is caused by a Mexican beer.

No. It can't xD. The survey is for beer drinkers. You can't take a sample size this small and try and pull it off as an entire nation, without even showing the statistical analysis. That is why I said demographics are important in all this.  For example, what was the average IQ of this sample size of beer drinkers?  For all we know they could've sampled 700 idiots who haven't a high school diploma. 

Again, I accept the flaws and inconsistencies small survey samples potentially have, but to completely mislead a viewing audience without considering the context of the survey, is unethical. That is the main issue. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

football forum
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...