Hope is not lost

While some say wisdom comes with age, younger Americans are better than their elders at separating factual from opinion statements in the news, according to a new analysis from Pew Research Center.

And the gap is noticeable:

About a third of 18- to 49-year-olds (32%) correctly identified all five of the factual statements as factual, compared with two-in-ten among those ages 50 and older. A similar pattern emerges for the opinion statements. Among 18- to 49-year-olds, 44% correctly identified all five opinion statements as opinions, compared with 26% among those ages 50 and older.

When looking at the 10 statements individually, younger adults were not only better overall at correctly identifying factual and opinion news statements – they could do so regardless of the ideological appeal of the statements.

Maybe social media aren’t so terrible after all.  And maybe cable TV is less than perfect?

That is from Jeffrey Gottfried and Elizabeth Grieco, via the excellent Kevin Lewis.


Well Duhhhh! Most of our problems are caused by people 50+ and especially those 60+ old boomers. I saw revoke people’s right to vote after 60. Come fight me!

Better yet, revoke the right to vote for all who post on the Internet using a pseudonym.

Just imagine the authors of the Federalist Papers today - think they would post their real name?

Anonymity has a lot of problems in the internet age, no question, and true anonymity is pretty close to impossible at this point anyways. Nonetheless, the idea that people need to be concretely identified before being allowed to participate in any activity is the sort of technical problem that our for profit surveillance society is working on solving, so as to improve the bottom line, if nothing else.

Actually Alan, we now have the disproof that "bad behavior needs anonymity."

When Twitter and Facebook brought millions of people online and encouraged them to use real names, many of them did, and then propagated the worst sorts of lies and hate.

It turns out there are thousands of people at least with nothing to lose by trying that to their true name.

Heck, it is true of some very prominent people using their true names.

Nice try Troll. You really are getting desperate, aren't you?

And there's some evidence that social media isn't polarizing voters (https://www.nytimes.com/2017/04/13/us/political-polarization-internet.html) -- though there is research in the opposite direction -- and that old people are the ones getting more polarized (https://www.forbes.com/sites/nextavenue/2017/04/16/polarization-is-growing-most-among-older-adults/#2f096a5e518e). Thanks for the debt and the polarization, grandma, grandpa, mom, and dad! I really appreciate it!

Do we know if such a change can be explained by increase in autistic nature? If so we might be in for a lot of the "the facts are true, the news is fake" (as Taleb headlined one of his articles)phenomenon.

The researchers strove to "include an equal number [of statements] that would appeal to the sensitivities of each side of the aisle". However, they tended to use senior-appealing statements that were opinions and young-appealing statements that were facts, thus biasing against seniors. It's harder for people to correctly identify appealing statements as opinions and unappealing statements as facts. Also, they count the first factual statement about Social Security as appealing to the Right, but that statement would not necessarily appeal to seniors, even conservative seniors.

The first 3 factual statements all seem to appeal to the young more than the old. The fourth might also as well (since liberals tend to have a lower opinion of the US health care system, especially in terms of cost). I don't know who finds the ISIS statement more appealing. So, 3-4 factual statements are probably easier for the young to classify correctly. The first 3 opinion statements appeal more to the old than the young. But, again, that actually *favors the young*. One might argue that the last two opinion statements are more appealing to the young (and hence more difficult for the young to correctly classify), but I thought I saw somewhere that the young don't actually support abortion rights more than the old. So, by my count 6-7 statements are easier for the young to correctly classify while only 0-3 statements might be easier for the old to correctly classify.

And do we know if there has actually been a change in how well people separate fact from opinion? I did this in class exercises in middle school English class back in the 90s. Beware of making assertions about trends based on a single data point. Maybe hope is not lost, but that may be because there was no hope to begin with.

I think you raise good points, but your conclusions go too far.

Pew's abortion poll numbers show young people being more pro-choice, although not by a large amount (63% for 18-29, 59% for 30-49, 56% for 50-64, 57% for 65+). Despite this slight bias, young people did significantly better. Young people are much more likely to support a $15 minimum wage, so that one definitely favors older people. I think the ISIS one favors older people too, but I could be wrong (the reasoning being that older people are far more likely to support Trump, and this question is essentially asking how effective his policies have been in that regard).

Also, I disagree that the democracy question would favor the young over the old. I don't think older people are more favorable to democracy. I'm skeptical of the social security/Medicare/Medicaid one as well - you say it "would not necessarily appeal to seniors", but then categorize it as appealing to young people. Making these changes would bring the numbers to 4-8 favoring young people, 2-6 favoring older people.

All in all, statements which favor older people tend to be about tied, statements which favor neither tend to have young people modestly ahead, and statements which favor young people tend to have young people far ahead. So there's good reason for optimism, just perhaps not as much as the article implies.

"I don't think older people are more favorable to democracy."

True, but the wording appeals to conservatives: "the greatest form of government" sounds very pro-American as it evokes the similar, "America is the greatest country in the world." "Democracy is the most inclusive form of government," or something similar would appeal more to the young.

Note also that, for the 4 statements I think obviously favor neither young nor old (last 2 facts and last 2 opinions), there is no statistically meaningful performance difference on 3 of them.

Yo. Fuck the olds!

Or, as the Margaret Atwood short story is titled, "Torch the Dusties"

TV news are great to inform people about hurricane warnings and not much else.

Watch a 30 min or 1 hour news. They are slow, at best they are the front page of an average journal. After 1 hour, you get informed about 7-8 topics at best. I've found that the amount of information conveyed TV news equals to ~10 min of reading a newspaper, twitter, online news.

TV news should be avoided. Its equivalent to mainlining a drug carefully engineered to provoke emotional response. Its remarkable how little information content is included. I stopped watching TV about 10 years ago, and don't miss it. And as you note, reading is much more efficient.

Even the weather channel can't bring themselves to just a factual weather report.

We do stream some TV shows. We play a game of identifying the "sermon of the week" embedded in the shows, although we have noticed lately that the sermon is increasingly a set piece with little to do with the plot line.

For any given question, the greatest difference is 12%. For each of these there appears to be heavy ideological divides between them. Further some of these are highly sensitive to how you parse them. E.g. Social Security, Medicare, and Medicaid make up the majority of the federal budget (i.e. "largest part") if we deem interest payments to not be part of the budget but not if we include interest + all other expenditures. It frankly is an opinion based on how you think one should view interest and the apportioning of the federal budget.

Controlling for education?

Agreed. More specifically, I was very happy when my daughter, in third and again in fourth and fifth grade here in the US, was taught the distinction between the two and trained to recognize them/ I had never seen that in France.

Maybe this was not taught even here two generations ago, and thus school, rather than social media, explains the better score of young people.

That being said, have you seen the questions? Just click on the link. They are trivial. Anything below 100% of correct answers can not induce optimism (is that a fact or an opinion?).

My (anecdote-based) impression with French people has been that they know things at such a pedantic level that every one of them trivially know the difference between the two. Of course, you certainly know better. But this is really the first time I am hearing of a positive aspect of US schooling that is absent in France.

Old people watch Fox News. Obsessively. Is it any wonder they can't distinguish fact from opinion. Why else would old people vote for Republicans: Fox News scares the geezers into believing Democrats are coming to take their Medicare and social security. That young people don't watch Fox News also explains their ability to distinguish facts from opinion. [As for the questions in the survey, I suspect many responses were actually based on whether the respondent agreed or disagreed with the statement rather than the distinction between fact and opinion.]

With not a word about the quality of the other news channels, your comment is vapid in the extreme

Generally, the ones that get the bombs are the good ones.

Or maybe their biases are just more in sync with the information being presented.

What's an 'opinion'? What about a question 'how many immigrants are there living in a state' - is it factual or opinion question? It depends on the definition of 'immigrant'. 'Most governments are inefficient and wasteful' - it depends on the definition on 'inefficient and wasteful', but it doesn't seem to be a type of question that cannot be answered factually.

"What about a question 'how many immigrants are there living in a state'"

I would say anything that only needs simple clarification is factual. "Do you mean born outside the state, or outside the country?"

"Most governments are inefficient and wasteful' - it depends on the definition on 'inefficient and wasteful', but it doesn't seem to be a type of question that cannot be answered factually."

Yes, that is why it is a statement of opinion and this survey classifies it as such.

Andy's definition of "factual" is purely opinion. The world is no black and white. I still can't comprehend whether light is a particle or a wave. I can admit it and I am actually proud of this fact.

Trying to make everyone so "certain" of the "facts" is a sign of disfunction.

Wave–particle duality is a good one, right at the limits of human comprehension, but that shouldn't mean we can't know that there are 12 eggs in dozen. Or yield to sophistry on simpler questions.

And there are 6.02 x 10 exp 23 in a mole.

approximately 6.02 x 10^23....don't think you go avogadros # factually correct.

As for particle duality...you might as well be talking about the three manifestations of the holy spirit as far as my hunter gatherer comprehension can get me. I am thankful that some of you higher IQ folks are here to look out for us, but I still have my doubts on some of this stuff and see quite a few people in the world who are way too certain about the "facts"

Also a bakers dozen is 13. So for both of even the simplest examples you can think of you are already way off. For Avogadro's #...all they have really been able to do is estimate the # down to like 10 digits...they get excited when they can add an extra digit of accuracy onto the estimate, but 6.02 x 10^23 is just a very rough estimate...off by billions of atoms at least.

I suprisingly found this article that summarises the idea: https://www.econlib.org/facts-opinions-and-the-pew-research-centers-pseudoscience/

So... are you over 40 and just trying to provide an anecdote to back up this research finding?

"And maybe cable TV is less than perfect"

Never has "maybe" been less necessary in a sentence.

Thinking back now to all the people who had all the reasons to ignore factcheck.org in these pages.

"Government is almost always wasteful and inefficient"

This is a factual statement.

What is your source for this statement.
For example, medicare is much more efficient than most insurance companies as is social security.

How efficient is the private sector in providing for the common defense?
Where government hires security forces as in Iraq, Afghanistan,etc the private forces are much more expensive than a comparable army or marine corp unit.

The government produces virtually no goods, virtually everything it does is a service. Does your data overweight goods vs services.

About the only thing private contractors does better is sanitation because the federal wage scale overpays at the bottom levels. That is why most government agencies hire private cleaning services rather than doing it itself -- the military and intelligence agencies go to a lot of trouble to put away all classified materials at the end of a shift because the private cleaners are not cleared to see it. see it.

Government actually build nothing. They hire private contractors to build roads and buildings. Why,because the firms specialize in these activities and it is their comparative advantage.

Can you show me an actual activity where government is more wasteful and inefficient than a private sector provider?

"For example, medicare is much more efficient than most insurance companies " No true. You need to account for the increase fraud allowed under medicare and things like no cost associated with revenue as the IRS handles that.

When I worked for the government, Al Gore was reinventing it, so we had to hire a bunch of consultants to do our jobs. They were without fail not very good at it, or maybe we weren't very good at teaching them how to do our jobs. It was also much more expensive, because they generally didn't seem so see their efforts as a job, but rather as a way to get a snout in the trough.

This was admittedly in an area where consultants wouldn't be very good, biological investigations of fish behavior where you have to swim around in cold water and it's counter-productive to spend a lot of time combing your hair afterward or figuring out the best coffee places on a new job-site or whether you were staying in the best motel in Podunk, far from your office tower, but even so they were pretty bad at it. I could hardly believe the billing statements as contrasted to the product. What? You charged out six guys for three weeks and you spent four hours in the water?

Of course one might wonder why the guvmint is swimming around looking at fish. But if it is, it's a lot more efficient to do it with a bunch of people who give a shit rather than whatever set of lazy boneheads a consultant might gather together for a research proposal.

Common defense is a great example of inefficiency and waste. The United States has natural borders and friendly neighbours and no enemies other than the ones it creates for itself. Yet it spends hundreds of billions that in no way contribute to the security of ordinary Americans and tries to pretend that countries with 1% of the military spending are some kind of threat.

The spending is very much "needed" in or to create justifications for the existence of the large military budget.

"Immigrants who are in the U.S. illegally have some rights under the constituion." But of course, it is right there in the constitution, the famous "illegal aliens rights" clause. Not. No, it was a Supreme Court OPINION that granted illegal aliens constitutional rights. In Reno v. Flores, Scalia wrote “it is well established that the Fifth Amendment entitles aliens to due process of law in deportation proceedings.” But that was his opinion. Given the anarchic nature of US jurisprudence, that opinion could be overturned at any time.

The decision is a fact, until it is overturned, and then it becomes another fact.

Jurisprudence is one of those areas that puts the lie to Daniel Patrick Moynihan's statement that "You are entitled to your own opinion, but you are not entitled to your own facts": for judges, their opinions are facts, and their position entitles them to it.

Fact, opinion, or stark raving:

"The Fake News Media, the true Enemy of the People, must stop the open & obvious hostility & report the news accurately & fairly."

(Let me guess, instead it is "TDS" to think high a official using "Enemy of the People" is a danger sign for the great American Republic. One week after bombs were mailed, by a guy who believed it.)

Pardon me, on the same day.

"Another suspicious package addressed to CNN has been intercepted. This time in Atlanta. All mail is being screened off site."


My 80 year old mother thinks Hannity is the news.

She's not the only one.

Interestingly, I could not find in the paper how they chose which statement was fact and which was opinion. I don't think this undermines the point too much, but still.

That Polling has some pretty bad biases. It's set up in a Leftwing normative stance. Enough to invalidate any conclusions.

At the very least, half the questions should be Rightwing normative. I suspect that it's a classic case of younger people being more conservative than older people and if you were to reverse the stance of the questions the results would vary.

Granted, that's pure speculation, but without extracting the obvious political bias to the polling, we can't answer the question allegedly asked.

"... younger people being more conservative than older people "

Should be "younger people being more left wing than older people"

Hope is not lost. Unless you're hoping for a media outlet to include an analysis of the null hypothesis. In which case, rest assured that hope has been smothered, and shot in the head for good measure.

"Maybe social media aren’t so terrible after all. And maybe cable TV is less than perfect?"

Maybe. Maybe not. But guesswork is certainly pointless and is generally harmful.

"ISIS lost a significant portion of its territory in Iraq and Syria in 2017"

This is not a factual statement, because it hinges entirely on the word "significant", which as a qualitative term is a matter of opinion. A more factual statement would be something like "ISIS no longer governs Al-Raqqa or Mosul", although maybe you could contest what the word "govern" means, rendering it a matter of opinion. I don't have too much of an issue with the other factual questions, certainly they aren't perfect, but it is concerning that Pew labels this one as a "factual" statement, and casts doubt on their judgment of what constitutes a "fact".

"In January 2015, ISIS militants controlled about 90,800 sq km, but by June 2017, that number dropped to 36,200, said IHS Markit."

Is this really something for which we have to quibble about the meaning of "significance?"

Yes, because the very point of this study is what constitutes a fact vs. an opinion: saying something is "significant" is an opinion, not a fact. The statement should have been framed the way you framed it if they wanted to make a factual statement.

I kind of get what you're saying, but I think you really did slide off a bit too far.

There are common sense, man on the street, kinds of facts that we should be able to agree on.

That more than half is significant is one of those things.

I also get what you are saying, but the idea that "common sense" or "consensus" decides what the facts are has been overturned again and again: a poll that attempts to highlight the difference between fact and opinion should not rely on "common sense" definitions of facts in their statements. Now, if this were a poll just about people's knowledge of world events? I wouldn't quibble with the use of "significant", but because of the specific context of this poll and its stated goals, I will quibble for a more rigorous definition of "fact" rather than the bare consensus definition.

For example, had ISIS lost over half its territory, but that territory was all desert and they still controlled Mosul, Raqqa, and other cities, people could very well argue that those losses were not "significant". This is the point I am making, the qualitative character of the word "significant", therefore inviting opinion.

We can all acknowledge the fact that the US government intelligence agencies have a lot of secrets that they want to keep. right? They also regularly prosecute employees(and former employees) that talk too much. Right?

On the ISIS thing it seems perfectly reasonable to use the population (of humans)under control as a bigger factor than the territory(square miles). I have no idea what the "factual" #'s are on population under ISIS control over the last two or three years....even then I can see A LOT of room for debate on what "control" means.

I still want to know more about those pictures I saw a couple years ago. The ones with all the brand new US made toyota trucks....it seems about a 35% chance that ISIS is just some sorta tool controlled by CIA esque elements of the US government(ala Emanuel Goldstein)...perhaps via the barbaric "ally" known as the Khashoggi killer House of Saud.

How are we supposed to know what the facts are around ISIS without being up close. The US government purposefully hides much of their activities in secretive black budget operations? also the US media has regularly cooperated in lying the US people in various wars. How do we know they aren't doing it right now?

It does certainly require some cultural literacy. If you ask if foboz lost fumbar in 2017 it does make much less sense.

It's a good idea to change "significant" to "substantial" every time you find yourself writing it. Unless you are in an elementary statistics class.

Or maybe older people are more likely to be moral realists who don't care about giving the answer they know that the experimenter is looking for, whereas younger people are more likely to either be moral relativists or care about giving the answer they know that the experimenter is looking for.

The article seems to conflate "factual statement" with "true statement".

I think the authors consider all the "factual" statements they list to also be true. But the respondents may not agree, and it's quite possible that the instructions they were given led them to think that they should not say a statement is "factual" if they think it is false, even though a factual statement does not become "opinion" just because it's false - it's just a false factual statement.

It's the *people in the study* that are conflating these words:

true factual
false opinion

From the example questions shown in the article's infographic all cases where older people underperformed younger people can be explained with a theory that the statements a conservative would be more likely to rate "true" (here also describing opinions that the reader agrees with) are categorized as Factual and those a conservative would rate "false" are categorized as Opinion at higher rates by the more-conservative older population.

Hopefully the study's instructions to participants used a more neutral phrasing like "Statements making claims about facts", though I expect even that wouldn't be enough to overcome the connotations from everyday usage of these words. The study shows pretty clearly that older persons less often know the dictionary definition of factual, but given that they also seem to equate opinion with "false" the study might still be showing a real issue with information-savvy among older persons.

Surely it's the responsibility of the people conducting the study to make VERY sure that the respondents know the meaning of what they are asking.

Although I wonder if the experimenters are themselves clear on what they're doing. Actually testing whether people can distinguish factual statements from opinion statements should surely involve giving them some factual statements that are clearly false (e.g., "slavery in the US was abolished more than 200 years ago").

As it is, they may not be testing whether people can distinguish factual statements from opinions, but rather whether they can distinguish whether (some) factual statements are true or false, or maybe whether they know the meanings of the words "factual" and "opinion".

"About a third of 18- to 49-year-olds (32%) correctly identified all five of the factual statements as factual, compared with two-in-ten among those ages 50 and older."
Couldn't this just indicate that the ability to discern decays with age.

Surely it's the responsibility of the people conducting the study to make VERY sure that the respondents know the meaning of what they are asking.

Although I wonder if the experimenters are themselves clear on what they're doing. Actually testing whether people can distinguish factual statements from opinion statements should surely involve giving them some factual statements that are clearly false (e.g., "slavery in the US was abolished more than 200 years ago").

As it is, they may not be testing whether people can distinguish factual statements from opinions, but rather whether they can distinguish whether (some) factual statements are true or false, or maybe whether they know the meanings of the words "factual" and "opinion".

Comments for this post are closed