|
Read the rules in the OP before posting, please.In order to ensure that this thread continues to meet TL standards and follows the proper guidelines, we will be enforcing the rules in the OP more strictly. Be sure to give them a re-read to refresh your memory! The vast majority of you are contributing in a healthy way, keep it up! NOTE: When providing a source, explain why you feel it is relevant and what purpose it adds to the discussion if it's not obvious. Also take note that unsubstantiated tweets/posts meant only to rekindle old arguments can result in a mod action. |
On March 09 2018 19:01 Acrofales wrote:Show nested quote +On March 09 2018 18:36 GreenHorizons wrote:On March 09 2018 17:33 WolfintheSheep wrote:On March 09 2018 17:02 GreenHorizons wrote:For the sake of keeping this relatively simple, explain to me (anyone) what it means when they say: Stories determined to be true rarely went wider than 1,000 people I feel like this has an obvious interpretation, but that doesn't make sense on it's face, so this must mean more than it's literal translation. What does it mean? It's actually a bit hard to dumb down, and I'm not sure I'm getting entirely what "rumour cascades" actually amount to. But lets say you heard a rumour from a "friend of a friend of a friend". Each friend is a person who is telling the story, as opposed to a person just retweeting. That 1,000 people is the number of people retweeting that one friend. I think the "cascade" is basically a person actively spreading the rumour (like a new tweet or discussion) as opposed to just a retweet. So a true rumour hits more than 1,000 people total, but each "cascade" rarely hits more than 1,000. And by percentage comparison, false rumours basically spread further and cascade more than true rumours. Which I guess is why this seems largely meaningless to me. It's basically useless for determining the depth and sincerity with which the root of the rumor true or false is held in public awareness. So a popular joke that ends up getting 'debunked' on snopes for clicks turns into evidence that false stories spread faster than true ones with a study like this. Satire will be picked up as people sharing 'false stories' in this study and more stuff like that. I mean Onion articles could be captured in this as "false stories" and them being more popular than a story about what actually happened being shared less (because they've heard that story on every other possible media source/platform), inferring that less people are aware of the truth than the lie that's a popular meme/joke on twitter. I mean I can think of basic/boring news stories that are far more widely known than some of the most viral twitter rumors, so something like "This marks the 18th year at war in a row" might not be as widely shared on twitter as "Obama personally executes Donald Trump in Guantanamo and replaces him with lizard clone" and the one that ends up on snopes is the backstory for the lizard clone, not confirmation that we are going on 18 years at 'war' with terrorism. I mean I'm far from being able to tell them how to get a clear picture math or variable wise, but this is generally pretty useless information without any potential practical application. Yes. Small steps. Understanding of complex phenomena hardly ever all comes at once in a giant "Eureka" moment. I'm sorry scientific inquiry is too slow for you. Maybe watch a cat video on youtube?
I think you misunderstand my perspective. I'm fine with the small steps. It's the whole (to succinctly sum it up) not mentioning in the article the part that says:
and the emotional reactions of recipients may be responsible for the differences observed.
Which is a fancy way of saying "we might be measuring jokes as sharing 'false stories'" In the article. I mean there's other aspects like that, but that example speaks to the theme pretty well.
You can find plenty of other examples in other articles that picked it up. I'd imagine I don't have to convince you about the problem of how news outlets report science.
|
On March 09 2018 19:19 GreenHorizons wrote:Show nested quote +On March 09 2018 19:01 Acrofales wrote:On March 09 2018 18:36 GreenHorizons wrote:On March 09 2018 17:33 WolfintheSheep wrote:On March 09 2018 17:02 GreenHorizons wrote:For the sake of keeping this relatively simple, explain to me (anyone) what it means when they say: Stories determined to be true rarely went wider than 1,000 people I feel like this has an obvious interpretation, but that doesn't make sense on it's face, so this must mean more than it's literal translation. What does it mean? It's actually a bit hard to dumb down, and I'm not sure I'm getting entirely what "rumour cascades" actually amount to. But lets say you heard a rumour from a "friend of a friend of a friend". Each friend is a person who is telling the story, as opposed to a person just retweeting. That 1,000 people is the number of people retweeting that one friend. I think the "cascade" is basically a person actively spreading the rumour (like a new tweet or discussion) as opposed to just a retweet. So a true rumour hits more than 1,000 people total, but each "cascade" rarely hits more than 1,000. And by percentage comparison, false rumours basically spread further and cascade more than true rumours. Which I guess is why this seems largely meaningless to me. It's basically useless for determining the depth and sincerity with which the root of the rumor true or false is held in public awareness. So a popular joke that ends up getting 'debunked' on snopes for clicks turns into evidence that false stories spread faster than true ones with a study like this. Satire will be picked up as people sharing 'false stories' in this study and more stuff like that. I mean Onion articles could be captured in this as "false stories" and them being more popular than a story about what actually happened being shared less (because they've heard that story on every other possible media source/platform), inferring that less people are aware of the truth than the lie that's a popular meme/joke on twitter. I mean I can think of basic/boring news stories that are far more widely known than some of the most viral twitter rumors, so something like "This marks the 18th year at war in a row" might not be as widely shared on twitter as "Obama personally executes Donald Trump in Guantanamo and replaces him with lizard clone" and the one that ends up on snopes is the backstory for the lizard clone, not confirmation that we are going on 18 years at 'war' with terrorism. I mean I'm far from being able to tell them how to get a clear picture math or variable wise, but this is generally pretty useless information without any potential practical application. Yes. Small steps. Understanding of complex phenomena hardly ever all comes at once in a giant "Eureka" moment. I'm sorry scientific inquiry is too slow for you. Maybe watch a cat video on youtube? I think you misunderstand my perspective. I'm fine with the small steps. It's the whole (to succinctly sum it up) not mentioning in the article the part that says: Show nested quote + and the emotional reactions of recipients may be responsible for the differences observed. Which is a fancy way of saying "we might be measuring jokes as sharing 'false stories'" In the article. I mean there's other aspects like that, but that example speaks to the theme pretty well. You can find plenty of other examples in other articles that picked it up. I'd imagine I don't have to convince you about the problem of how news outlets report science.
Now I read the ABC article too, and that line is there, verbatim. In fact, in this case I have no quibble with the popular press version of this scientific article. Even the title doesn't overstate it.
And as for your point about jokes, if they advance to the point where they are fact checked, then Poe's law has gotten hold of them and ppl are misrepresenting it as a "rumor" rather than satire/joke/meme. As was the case for things like the Colbert report and various articles on the Onion...
|
On March 09 2018 19:34 Acrofales wrote:Show nested quote +On March 09 2018 19:19 GreenHorizons wrote:On March 09 2018 19:01 Acrofales wrote:On March 09 2018 18:36 GreenHorizons wrote:On March 09 2018 17:33 WolfintheSheep wrote:On March 09 2018 17:02 GreenHorizons wrote:For the sake of keeping this relatively simple, explain to me (anyone) what it means when they say: Stories determined to be true rarely went wider than 1,000 people I feel like this has an obvious interpretation, but that doesn't make sense on it's face, so this must mean more than it's literal translation. What does it mean? It's actually a bit hard to dumb down, and I'm not sure I'm getting entirely what "rumour cascades" actually amount to. But lets say you heard a rumour from a "friend of a friend of a friend". Each friend is a person who is telling the story, as opposed to a person just retweeting. That 1,000 people is the number of people retweeting that one friend. I think the "cascade" is basically a person actively spreading the rumour (like a new tweet or discussion) as opposed to just a retweet. So a true rumour hits more than 1,000 people total, but each "cascade" rarely hits more than 1,000. And by percentage comparison, false rumours basically spread further and cascade more than true rumours. Which I guess is why this seems largely meaningless to me. It's basically useless for determining the depth and sincerity with which the root of the rumor true or false is held in public awareness. So a popular joke that ends up getting 'debunked' on snopes for clicks turns into evidence that false stories spread faster than true ones with a study like this. Satire will be picked up as people sharing 'false stories' in this study and more stuff like that. I mean Onion articles could be captured in this as "false stories" and them being more popular than a story about what actually happened being shared less (because they've heard that story on every other possible media source/platform), inferring that less people are aware of the truth than the lie that's a popular meme/joke on twitter. I mean I can think of basic/boring news stories that are far more widely known than some of the most viral twitter rumors, so something like "This marks the 18th year at war in a row" might not be as widely shared on twitter as "Obama personally executes Donald Trump in Guantanamo and replaces him with lizard clone" and the one that ends up on snopes is the backstory for the lizard clone, not confirmation that we are going on 18 years at 'war' with terrorism. I mean I'm far from being able to tell them how to get a clear picture math or variable wise, but this is generally pretty useless information without any potential practical application. Yes. Small steps. Understanding of complex phenomena hardly ever all comes at once in a giant "Eureka" moment. I'm sorry scientific inquiry is too slow for you. Maybe watch a cat video on youtube? I think you misunderstand my perspective. I'm fine with the small steps. It's the whole (to succinctly sum it up) not mentioning in the article the part that says: and the emotional reactions of recipients may be responsible for the differences observed. Which is a fancy way of saying "we might be measuring jokes as sharing 'false stories'" In the article. I mean there's other aspects like that, but that example speaks to the theme pretty well. You can find plenty of other examples in other articles that picked it up. I'd imagine I don't have to convince you about the problem of how news outlets report science. Now I read the ABC article too, and that line is there, verbatim. In fact, in this case I have no quibble with the popular press version of this scientific article. Even the title doesn't overstate it. And as for your point about jokes, if they advance to the point where they are fact checked, then Poe's law has gotten hold of them and ppl are misrepresenting it as a "rumor" rather than satire/joke/meme. As was the case for things like the Colbert report and various articles on the Onion...
That's what I get for not control f'ing.
Fair enough, still doesn't say it in a way I think most people sharing the headline would interpret to mean what it means. I'll give them credit for putting it in there even if they didn't explain what that means about the headline conclusion at all.
As to the part of jokes, I think that still presents a pretty big problem with understanding and drawing conclusions from the data.
EDIT: I take your other points, but could you explain the whole "Stories determined to be true rarely went wider than 1,000 people" means?
Do they literally mean that true stories are seen less than stories that are false, or that despite more people sharing the false rumors true stories are seen by more people on twitter, or do they not address the difference between 10,000 tweets seen by 10 people each and 1 tweet seen by 1,000,000 people without any of them sharing it?
That is to say, if CNN shared a story on twitter that got 100 retweets but 10,000,000 impressions, is that less 'shared' than a fake rumor that had 10,000 tweets but less than 1,000,000 impressions?
If so I don't know if saying it spreads more/faster is fully accurate, it spreads more organically and from the bottom up (as they say both true and false stories are amplified by bots equally). But CNN showing 10,000,000 people on twitter something is more widely shared than 10,000 people showing 5 of their friends something imo and I'm not able to tell if the study addresses that.
EDIT2: Was there any specific examples of widely shared fake or narrowly shared real stories hidden in there? I feel like that could give us some valuable information as well.
|
On March 09 2018 14:33 Womwomwom wrote:
ESA members generally treat this sort of video violence with some degree of care. Like using appropriate YouTube thumbnails and gating the content behind an age wall and a whole bunch of other warnings.
I tried checking but it didn’t seem like the Trump Administration put an age wall behind that YouTube video. The thumbnail was also of some guy getting their brains blown out in Sniper Elite. As usual, they went in guns blazing only to do a whole lot of collateral damage because they didn’t use their collective brains.
And yes, there was definitely a pushback against horror movies. I dunno about the US but there was a period in UK cinema that the authorities went after so called video nasties.
Fucking Mary Whitehouse. We're still feeling the cultural effects of her nonsense.
|
This 'fake news spreads faster' story..
Does it strike anyone else as stating the obvious?
I mean I could make up a story now that more people would be interested in than a retelling of the last 45 minutes of my life.
Wild, bizarre and creative stories are more fun.
|
Fiction is more compelling to read about than reality. It’s sort of sad it took us all this long to see this flaw with social media.
|
On March 09 2018 20:18 Plansix wrote:
Fiction is more compelling to read about than reality. It’s sort of sad it took us all this long to see this flaw with social media.
Luckily the US president is actively fighting this.
Make the reality more like a badly written cartoon so it doesn't get outperformed on the entertainment scale.
|
Did anyone stumble on any examples of the rumors they looked at and how they compared (like top 10 shared true vs false stories)?
|
On March 09 2018 20:14 Jockmcplop wrote:
This 'fake news spreads faster' story..
Does it strike anyone else as stating the obvious?
I mean I could make up a story now that more people would be interested in than a retelling of the last 45 minutes of my life.
Wild, bizarre and creative stories are more fun.
Sure, but there is a place for real, hard-hitting journalism as well. I think there is a case for a little bit more realistic view of the world being the "conventional view." For example, I think that it would be beneficial for both countries if there were more free trade that occurred between the US & Canada. They are technically our closest neighbors, & they do trade the most with us, although we do technically trade the most with 1) China, 2) Japan, 3) them. So anyways this article is pushing a narrative or promoting a specific agenda item, but it is based on factual evidence and I think there is a case for this being a good idea. More legal immigration allowed & encouraged would be good as well.
https://www.nationalreview.com/2017/05/cross-border-us-canada-commerce-make-it-easier-shred-or-revise-nafta/
I guess what I'm saying is that even though "fake news spreads faster" through social media & papers like USA Today, definitely "real news is better"
|
On March 09 2018 19:44 GreenHorizons wrote:Show nested quote +On March 09 2018 19:34 Acrofales wrote:On March 09 2018 19:19 GreenHorizons wrote:On March 09 2018 19:01 Acrofales wrote:On March 09 2018 18:36 GreenHorizons wrote:On March 09 2018 17:33 WolfintheSheep wrote:On March 09 2018 17:02 GreenHorizons wrote:For the sake of keeping this relatively simple, explain to me (anyone) what it means when they say: Stories determined to be true rarely went wider than 1,000 people I feel like this has an obvious interpretation, but that doesn't make sense on it's face, so this must mean more than it's literal translation. What does it mean? It's actually a bit hard to dumb down, and I'm not sure I'm getting entirely what "rumour cascades" actually amount to. But lets say you heard a rumour from a "friend of a friend of a friend". Each friend is a person who is telling the story, as opposed to a person just retweeting. That 1,000 people is the number of people retweeting that one friend. I think the "cascade" is basically a person actively spreading the rumour (like a new tweet or discussion) as opposed to just a retweet. So a true rumour hits more than 1,000 people total, but each "cascade" rarely hits more than 1,000. And by percentage comparison, false rumours basically spread further and cascade more than true rumours. Which I guess is why this seems largely meaningless to me. It's basically useless for determining the depth and sincerity with which the root of the rumor true or false is held in public awareness. So a popular joke that ends up getting 'debunked' on snopes for clicks turns into evidence that false stories spread faster than true ones with a study like this. Satire will be picked up as people sharing 'false stories' in this study and more stuff like that. I mean Onion articles could be captured in this as "false stories" and them being more popular than a story about what actually happened being shared less (because they've heard that story on every other possible media source/platform), inferring that less people are aware of the truth than the lie that's a popular meme/joke on twitter. I mean I can think of basic/boring news stories that are far more widely known than some of the most viral twitter rumors, so something like "This marks the 18th year at war in a row" might not be as widely shared on twitter as "Obama personally executes Donald Trump in Guantanamo and replaces him with lizard clone" and the one that ends up on snopes is the backstory for the lizard clone, not confirmation that we are going on 18 years at 'war' with terrorism. I mean I'm far from being able to tell them how to get a clear picture math or variable wise, but this is generally pretty useless information without any potential practical application. Yes. Small steps. Understanding of complex phenomena hardly ever all comes at once in a giant "Eureka" moment. I'm sorry scientific inquiry is too slow for you. Maybe watch a cat video on youtube? I think you misunderstand my perspective. I'm fine with the small steps. It's the whole (to succinctly sum it up) not mentioning in the article the part that says: and the emotional reactions of recipients may be responsible for the differences observed. Which is a fancy way of saying "we might be measuring jokes as sharing 'false stories'" In the article. I mean there's other aspects like that, but that example speaks to the theme pretty well. You can find plenty of other examples in other articles that picked it up. I'd imagine I don't have to convince you about the problem of how news outlets report science. Now I read the ABC article too, and that line is there, verbatim. In fact, in this case I have no quibble with the popular press version of this scientific article. Even the title doesn't overstate it. And as for your point about jokes, if they advance to the point where they are fact checked, then Poe's law has gotten hold of them and ppl are misrepresenting it as a "rumor" rather than satire/joke/meme. As was the case for things like the Colbert report and various articles on the Onion... That's what I get for not control f'ing. Fair enough, still doesn't say it in a way I think most people sharing the headline would interpret to mean what it means. I'll give them credit for putting it in there even if they didn't explain what that means about the headline conclusion at all. As to the part of jokes, I think that still presents a pretty big problem with understanding and drawing conclusions from the data. EDIT: I take your other points, but could you explain the whole "Stories determined to be true rarely went wider than 1,000 people" means? Do they literally mean that true stories are seen less than stories that are false, or that despite more people sharing the false rumors true stories are seen by more people on twitter, or do they not address the difference between 10,000 tweets seen by 10 people each and 1 tweet seen by 1,000,000 people without any of them sharing it? That is to say, if CNN shared a story on twitter that got 100 retweets but 10,000,000 impressions, is that less 'shared' than a fake rumor that had 10,000 tweets but less than 1,000,000 impressions? If so I don't know if saying it spreads more/faster is fully accurate, it spreads more organically and from the bottom up (as they say both true and false stories are amplified by bots equally). But CNN showing 10,000,000 people on twitter something is more widely shared than 10,000 people showing 5 of their friends something imo and I'm not able to tell if the study addresses that. EDIT2: Was there any specific examples of widely shared fake or narrowly shared real stories hidden in there? I feel like that could give us some valuable information as well.
First off, they do not take views, or interpretations (such as believing) into account and look specifically at how rumours spread. As you said, you could propagate something you *know* is false purely for entertainment value. However, there are other studies on *why* people choose to propagate some rumors but not others. This is only about the statistics of rumors spreading.
This also means it doesn't say anything about "impressions". I don't think Twitter even has that data, but if they do, it wasn't used in this study (and I have to say, I haven't clicked through to study the dataset myself, although it is freely available for anybody who wants to look. So they do not say anything about views, and I don't know in how far retweets correlate with views... research into facebook likes or even shares is rather hazy on whether likes translate to interest, and that is the most studied social network, so basically, the jury is still out on this one.
However, your question is phrased in a way that they do not answer, but if you ask it the other way round, they do answer it. The question they answer isn't whether a piece of news you're sharing is more likely to be true or false. They answer whether you are more likely to share a false piece of news than a true one. They do not say anything at all about the statistics of whether any given rumor you see on twitter is more likely to be true or false, as that would depend not only on the likelihood of retweeting but also on the number of initial rumors of either kind. You can probably find the statistics for this in their dataset.
Now keep that in mind when reading "stories determined to be true rarely went wider than 1,000 people". It doesn't mean that if a story went wider than 1,000 people it is more likely to be false. It only means that of the true stories in their data set, a very small percentage of them went wider than 1,000 people. Exactly how few is "very few" you'd have to really dig into their supplementary material, which gives the detailed analysis: http://science.sciencemag.org/content/sci/suppl/2018/03/07/359.6380.1146.DC1/aap9559_Vosoughi_SM.pdf
I didn't see any specific rumors they talk about, but there is a quantitive analysis surrounding events in Graph E:
+ Show Spoiler +
@Jockmcplop: to some extent it is stating the obvious. However, this is the first real statistical analysis of the kind that identifies that Aunty Ethel is more likely to share the juicy gossip that she heard from her close friend Judy who said she read it in the gossip column while waiting to get her hair done, than that same Aunty Ethel is to share the boring report about what bill congress passed today, despite even Aunty Ethel knowing that the former is likely false and the latter likely true. And at the end of the day, if anybody gets their news exclusively from Aunty Ethel, they'll get a very distorted picture of what is actually going on in the world.
And that was quite okay when we knew aunty Ethel personally and took what she said with a grain of salt, but is a lot more complicated when the source is anonymous, and all the online ranking systems are designed to use "retweets" to be a measure of trustworthiness of the news article.
|
On March 09 2018 20:31 GreenHorizons wrote:
Did anyone stumble on any examples of the rumors they looked at and how they compared (like top 10 shared true vs false stories)?
There aren't any examples, but here's the relevant bit of the methodology
Here we investigate the differential diffusion of true, false, and mixed (partially true, partially false) news stories using a comprehensive data set of all of the fact-checked rumor cascades that spread on Twitter from its inception in 2006 to 2017. The data include ~126,000 rumor cascades spread by ~3 million people more than 4.5 million times. We sampled all rumor cascades investigated by six independent fact-checking organizations (snopes.com, politifact.com, factcheck.org, truthorfiction.com, hoax-slayer.com, and urbanlegends. about.com) by parsing the title, body, and verdict (true, false, or mixed) of each rumor investigation reported on their websites and automatically collecting the cascades corresponding to those rumors on Twitter.
The result was a sample of rumor cascades whose veracity had been agreed on by these organizations between 95 and 98% of the time. We cataloged the diffusion of the rumor cascades by collecting all English-language replies to tweets that contained a link to any of the aforementioned websites from 2006 to 2017 and used optical character recognition to extract text from images where needed. For each reply tweet, we extracted the original tweet being replied to and all the retweets of the original tweet. Each retweet cascade represents a rumor propagating on Twitter that has been verified as true or false by the fact-checking organizations (see the supplementary materials for more details on cascade construction).
We then quantified the cascades’ depth (the number of retweet hops from the origin tweet over time, where a hop is a retweet by a new unique user), size (the number of users involved in the cascade over time), maximum breadth (the maximum number of users involved in the cascade at any depth), and structural virality (23) (a measure that interpolates between content spread through a single, large broadcast and that which spreads through multiple generations, with any one individual directly responsible for only a fraction of the total spread)
In short, of all the stories deemed true or false by fact-checking sites, the ones deemed false spread faster and reached more users on twitter.
The problem with concluding from this that false news spreads faster than true news is that the study only analyzes rumours and claims. Look for example at the biggest news stories from yesterday: tariff announcement, Skirpal's assassination and the SK delegation visit. Why would any of these stories be on fact checking websites? There is nothing they can check about them. And of course these stories reached more people than "London Closes 500 Churches and Opens 423 New Mosques", something that snopes deemed false yesterday.
|
On March 09 2018 21:38 Dan HH wrote:Show nested quote +On March 09 2018 20:31 GreenHorizons wrote:
Did anyone stumble on any examples of the rumors they looked at and how they compared (like top 10 shared true vs false stories)? There aren't any examples, but here's the relevant bit of the methodology Show nested quote +Here we investigate the differential diffusion of true, false, and mixed (partially true, partially false) news stories using a comprehensive data set of all of the fact-checked rumor cascades that spread on Twitter from its inception in 2006 to 2017. The data include ~126,000 rumor cascades spread by ~3 million people more than 4.5 million times. We sampled all rumor cascades investigated by six independent fact-checking organizations (snopes.com, politifact.com, factcheck.org, truthorfiction.com, hoax-slayer.com, and urbanlegends. about.com) by parsing the title, body, and verdict (true, false, or mixed) of each rumor investigation reported on their websites and automatically collecting the cascades corresponding to those rumors on Twitter.
The result was a sample of rumor cascades whose veracity had been agreed on by these organizations between 95 and 98% of the time. We cataloged the diffusion of the rumor cascades by collecting all English-language replies to tweets that contained a link to any of the aforementioned websites from 2006 to 2017 and used optical character recognition to extract text from images where needed. For each reply tweet, we extracted the original tweet being replied to and all the retweets of the original tweet. Each retweet cascade represents a rumor propagating on Twitter that has been verified as true or false by the fact-checking organizations (see the supplementary materials for more details on cascade construction).
We then quantified the cascades’ depth (the number of retweet hops from the origin tweet over time, where a hop is a retweet by a new unique user), size (the number of users involved in the cascade over time), maximum breadth (the maximum number of users involved in the cascade at any depth), and structural virality (23) (a measure that interpolates between content spread through a single, large broadcast and that which spreads through multiple generations, with any one individual directly responsible for only a fraction of the total spread) In short, of all the stories deemed true or false by fact-checking sites, the ones deemed false spread faster and reached more users on twitter. The problem with concluding from this that false news spreads faster than true news is that the study only analyzes rumours and claims. Look for example at the biggest news stories from yesterday: tariff announcement, Skirpal's assassination and the SK delegation visit. Why would any of these stories be on fact checking websites? There is nothing they can check about them. And of course these stories reached more people than "London Closes 500 Churches and Opens 423 New Mosques", something that snopes deemed false yesterday.
Not entirely true. They controlled for that bias by having some undergrads fact check a random sample of news articles that didn't appear on fact checkers. And then they ran the same tests on that new dataset with similar findings.
|
|
|
On March 09 2018 21:43 Acrofales wrote:Show nested quote +On March 09 2018 21:38 Dan HH wrote:On March 09 2018 20:31 GreenHorizons wrote:
Did anyone stumble on any examples of the rumors they looked at and how they compared (like top 10 shared true vs false stories)? There aren't any examples, but here's the relevant bit of the methodology Here we investigate the differential diffusion of true, false, and mixed (partially true, partially false) news stories using a comprehensive data set of all of the fact-checked rumor cascades that spread on Twitter from its inception in 2006 to 2017. The data include ~126,000 rumor cascades spread by ~3 million people more than 4.5 million times. We sampled all rumor cascades investigated by six independent fact-checking organizations (snopes.com, politifact.com, factcheck.org, truthorfiction.com, hoax-slayer.com, and urbanlegends. about.com) by parsing the title, body, and verdict (true, false, or mixed) of each rumor investigation reported on their websites and automatically collecting the cascades corresponding to those rumors on Twitter.
The result was a sample of rumor cascades whose veracity had been agreed on by these organizations between 95 and 98% of the time. We cataloged the diffusion of the rumor cascades by collecting all English-language replies to tweets that contained a link to any of the aforementioned websites from 2006 to 2017 and used optical character recognition to extract text from images where needed. For each reply tweet, we extracted the original tweet being replied to and all the retweets of the original tweet. Each retweet cascade represents a rumor propagating on Twitter that has been verified as true or false by the fact-checking organizations (see the supplementary materials for more details on cascade construction).
We then quantified the cascades’ depth (the number of retweet hops from the origin tweet over time, where a hop is a retweet by a new unique user), size (the number of users involved in the cascade over time), maximum breadth (the maximum number of users involved in the cascade at any depth), and structural virality (23) (a measure that interpolates between content spread through a single, large broadcast and that which spreads through multiple generations, with any one individual directly responsible for only a fraction of the total spread) In short, of all the stories deemed true or false by fact-checking sites, the ones deemed false spread faster and reached more users on twitter. The problem with concluding from this that false news spreads faster than true news is that the study only analyzes rumours and claims. Look for example at the biggest news stories from yesterday: tariff announcement, Skirpal's assassination and the SK delegation visit. Why would any of these stories be on fact checking websites? There is nothing they can check about them. And of course these stories reached more people than "London Closes 500 Churches and Opens 423 New Mosques", something that snopes deemed false yesterday. Not entirely true. They controlled for that bias by having some undergrads fact check a random sample of news articles that didn't appear on fact checkers. And then they ran the same tests on that new dataset with similar findings.
But they still only looked at rumours. I'm not denying that false rumours spread faster than true ones, my main issue with this study is that it uses the word news in the title instead of rumours.
We then asked the student annotators to investigate these rumors. Some of what the system had detected as rumors ended up not being rumors (false positives), or were already investigated by one of the fact checking organizations; these were discarded (131 were discarded)
|
|
|
One word that I would not use for Trump's administration is "paralyzed." The problem with the daily barrage of "Trump's government is dysfunctional" stories in the media is that his government has been very aggressively pushing his agenda forward on multiple fronts, notwithstanding some high-profile failures involving the legislature such as the Obamacare repeal.
|
I believe the argument is that agencies like the EPA and VA are unable to fulfill their roles due to Trump’s appointees actively trying to dismantle the agencies from within. We just need to look at the State Department and EPA for the best examples.
Also, before people start talking about unelected bureaucrats, remember that most of these civil servants are prohibited from defending themselves by going on cable TV and complaining about how their agency is run. Their director and other people trying to dismantle these agencies are not. These are easy political targets because they can’t defend themselves and congress licks the president’s boot.
|
On March 10 2018 01:47 xDaunt wrote:One word that I would not use for Trump's administration is "paralyzed." The problem with the daily barrage of "Trump's government is dysfunctional" stories in the media is that his government has been very aggressively pushing his agenda forward on multiple fronts, notwithstanding some high-profile failures involving the legislature such as the Obamacare repeal.
You don't think the Trump administration is dysfunctional?
Like, really?
|
On March 10 2018 01:47 xDaunt wrote:One word that I would not use for Trump's administration is "paralyzed." The problem with the daily barrage of "Trump's government is dysfunctional" stories in the media is that his government has been very aggressively pushing his agenda forward on multiple fronts, notwithstanding some high-profile failures involving the legislature such as the Obamacare repeal.
You are focusing on messaging. Other people are focusing on government's ability to accomplish tasks and generally keep the lights on.
|
On March 10 2018 02:24 Mohdoo wrote:Show nested quote +On March 10 2018 01:47 xDaunt wrote:One word that I would not use for Trump's administration is "paralyzed." The problem with the daily barrage of "Trump's government is dysfunctional" stories in the media is that his government has been very aggressively pushing his agenda forward on multiple fronts, notwithstanding some high-profile failures involving the legislature such as the Obamacare repeal. You are focusing on messaging. Other people are focusing on government's ability to accomplish tasks and generally keep the lights on.
You're talking to someone whose ideal form of government is one that doesn't have lights. Or anything else, except a very big army.
|
|
|
|
|
|
EPT![[image loading]](https://d2ufo47lrtsv5s.cloudfront.net/content/sci/359/6380/1146/F1.large.jpg)
