Even NPR has joined the MSM race flaming

NPR, like all other media outlets, has an agenda.

They also know that increased listenership = increased ad revenue. The Teeming Millions (for the most part) lick their lips when the prospect of drama presents itself; sensationalized reporting is the norm because everybody is doing it, and NPR must do the same to remain competitive in their market share.

Everyone runs home Mama Dollar in the media business.

originally posted by: Asktheanimals
FEMA flood insurance came about because rich people were losing their vacation homes along the coast. People who could afford to rebuild it themselves got off the hook. Before beach vacation homes came of vogue in the 70's real estate along the coast was dirt cheap and so were the houses. Everybody knew that in 20 to 30 years a storm would take it they would have to rebuild. We can thank FEMA for making beach vacations unaffordable to many Americans by skewing property values.

John Stossels did a piece on FEMA flood insurance I'll have to find.
He lost his vacation home to a hurricane and FEMA paid him to replace it.

As for NPR I haven't trusted them in over 20 years. Their minority focus on everything is so predictable they've become a cliche.

Yup, government should not be subsidizing people to build homes in known frequent disaster areas. I have no problem with people who want beachfront or flood prone property, but government should not be insuring it. If you can't pay out of pocket to replace the home or the private insurance market won't cover it, then maybe you need to reconsider buying the home.

a reply to: Edumakated

When I was in J-school many long moons ago, there was a class about stats for journalism, but it was all about how to generate stats specifically to say what you want, not about how to read stats in general.

So I wouldn't assume they don't know anything about stats. They may simply have created the numbers they needed for their purposes.

originally posted by: ketsuko
a reply to: Edumakated

When I was in J-school many long moons ago, there was a class about stats for journalism, but it was all about how to generate stats specifically to say what you want, not about how to read stats in general.

So I wouldn't assume they don't know anything about stats. They may simply have created the numbers they needed for their purposes.

I don't know how anyone could have been taught basic statistics and not question the validity of the data as being presented...

A big part of learning statistics is understanding the flaws and interpreting data so you are not mislead... at least that is how we were taught in my undergrad business and MBA courses.

Except like New Orleans that’s where much of the affordable housing is, hence low income people sometimes have to live there, hence disproportionally affected. This is where the intersectionality claims come in, because a higher % of poc are poor, more are in some of these zones.

a reply to: neo96

Sure, but most people never have beyond a basic intro stats class, if at all. Those usually doesn’t get into advanced data models and scientific based research.

MBA schools take it a step up, as did my MPA, and as do most quantitative and research focused phds.
a reply to: Edumakated

Well, having taken one semester of honors undergrad stats, and three at grad school level, I can say that unless engages one in that level plus relevant data analysis scientifically or professionally, they won’t be a data scientist nor statistician.

One semester won’t cut it unless someone also does a ton of relevant professional work.

I’ll also say that even that’s not enough to professionally analyze a range of studies. One also needs work or study on a range of research methodology, such as physical or social sciences. There are concepts and designs in that which I didn’t encounter in my stats/data coursework, although they are supposed to be mutually supporting.

Sure a semester will introduce the basics.
a reply to: ketsuko

originally posted by: Quetzalcoatl14
Except like New Orleans that’s where much of the affordable housing is, hence low income people sometimes have to live there, hence disproportionally affected. This is where the intersectionality claims come in, because a higher % of poc are poor, more are in some of these zones.

a reply to: neo96

No they are not disproportionately affected. If they don't buy the insurance they don't get the coverage, period.
It doesn't matter what color they are.

Now if you show me a study where a group of latino, or black folks buy the insurance, and a group of white people do the same. Then the government comes in and only rebuilds the more of the white people houses, now that right there is racist/discriminatory.

Btw. I was a professional bean counter for many years. You can manipulate data to say anything you want by leaving out some pesky details...
I worked for a large tech company, and also a bank....believe me, we did it on a regular basis...

a reply to: Edumakated

The course in question was about how to set up samples and whom to sample, etc., and how to present the numbers to create the impression you wanted. In short, the numbers don't lie, but they may not exactly be telling the whole truth either.

A good example is the 97% consensus in climate science. It's a misleading stat not because they didn't find that number, but because of how they set up the data set in question to arrive at it.

I’m a data analyst and former researcher, not going to impress me with that one.

Of course one can, but obviously that’s an abuse of research and stats.

My point was a direct response to neo’s comment ignoring how poverty factually often congregates in such locations. Please address that point, if choosing to respond to me.


a reply to: JAGStorm

That would be either a research or data error/manipulation, not a truth that “statistics lie.” They do when done incorrectly and misrepresented or not properly qualified, which is what a good researcher is supposed to do.

a reply to: ketsuko

originally posted by: Quetzalcoatl14
That would be either a research or data error/manipulation, not a truth that “statistics lie.” They do when done incorrectly and misrepresented or not properly qualified, which is what a good researcher is supposed to do.

a reply to: ketsuko

Basically in that case they had to select a bunch of different qualifiers to arrive at a data set that would give them their 97% number, so the original data set was heavily manipulated. However, the media ran with the number without mentioning that part of the study.

The class in question was all about how to do this, and this is why polling in the press should always be viewed with a very jaundiced eye. They know how to write the questions and how to pick the area to poll with those questions to have the best chance to massaging their outcomes into something they want.