←back to thread

CDC data are disappearing

(www.theatlantic.com)
749 points doener | 4 comments | 01 Feb 25 11:52 UTC | HN request time: 0s | source
Show context
breadwinner ◴[01 Feb 25 20:59 UTC] No.42902252[source]▶
Data is the ultimate Fact Check. This is a President that's adamantly opposed to fact checking [1] and has even coerced Facebook to drop fact checking. Of course they don't want data on government sites that disprove their "alternate facts".

[1] https://thehill.com/homenews/campaign/4920827-60-minutes-tru...

replies(10): >>42902356 #>>42902413 #>>42902434 #>>42902630 #>>42902793 #>>42902978 #>>42903439 #>>42903684 #>>42904050 #>>42918244 #
cle ◴[01 Feb 25 22:15 UTC] No.42902978[source]▶
> Data is the ultimate Fact Check.

This is wrong IMO. Data can be missing, incomplete, biased, skewed, and even just plain wrong. Cherry-picked data can be worse than no data.

The ultimate fact check is a scientific process of collecting data, modeling it, scrutinizing it and its methodology and the entities involved, contextualizing it, cross-checking, replicating, etc.

What media likes to call "fact checking" to me feels more motivated by punchy headlines and chyrons.

replies(4): >>42903019 #>>42903435 #>>42905930 #>>42911802 #
breadwinner ◴[01 Feb 25 22:18 UTC] No.42903019[source]▶
> Data can be missing, incomplete, biased, skewed, and even just plain wrong.

All true of course. The solution for that is more data, not less.

replies(2): >>42903042 #>>42905140 #
1. _heimdall ◴[02 Feb 25 02:38 UTC] No.42905140[source]▶
Data is always lacking. More data may help you be more confident in your conclusion, but it will never be certain.
replies(1): >>42911828 #
2. johnnyanmac ◴[02 Feb 25 21:06 UTC] No.42911828[source]▶
Many of us here are engineers. Similar to our work, databcollected by scientists will eventually get to "good enough" in that interpretations are nearly indistinguishable from the truth. We don't need to understand every atom in the atmosphere to predict rain coming soon, for example. We don't need to do a full body scan to see visible breast cancer lumps.
replies(1): >>42912620 #
3. _heimdall ◴[02 Feb 25 22:41 UTC] No.42912620[source]▶
I agree with you, but that wasn't my point. The post I replied to simply said the answer is more data. Without any more context about what kind of answer it is or how it should be used, it seemed important to me to remind whoever passes by that data alone does not make truth and its always worth keeping in mind that what we "know" today may be considered false tomorrow.

What you're talking about is more a question of scape, impact, and how accurate a prediction really needs to be. Of course we don't need to measure every atom to predict the weather - weather predictions are wrong all the time and rarely is that more than an inconvenience.

replies(1): >>42912718 #
4. johnnyanmac ◴[02 Feb 25 22:53 UTC] No.42912718{3}[source]▶
Naturally, garbage in, garbage out. Anyone who's worked any job should under stand that. And no data is ever perfect when measuring nature.

But I'm giving a best faith interpretation that the ones collecting the data are competent and have goals on what the data is collected for. We have too much talent flowing to assume the worst. We'll see how the next 4 years challenges my assumptions, though.

>What you're talking about is more a question of scape, impact, and how accurate a prediction really needs to be

Yes. That goal of data is to approximate the truth. More (good) data helps those who can interpret it to make better guesses. So the base truth of "we need more data then" is true. With a good faith interpretation.