Better Informed?

The Internet has made Americans feel better informed, according to a new Pew study. The key point -- 76 percent of respondents say the Internet has made "average Americans" better informed, and 77 percent say the net has made "America's students" better informed. You can see the raw questionnaire here.

A couple of points.

• First, this is perception of being informed, not actual knowledge. Perceived knowledge and actual knowledge are positively correlated, but not perfectly so, and there are segments of the population who believe they are better informed than they actually are, and the consumption of certain media can exacerbate this effect (i.e., talk radio). Here respondents were asked not whether they are better informed thanks to the Internet, but rather whether others are better informed. That's a subtle difference.

• There's an interesting table on the types of information available on the net that makes you feel better informed. Look at it (Q34 on the raw questionnaire, but page 2 of the pdf). In most cases people feel better informed, but the fascinating failure, if you will, is for "civic life and government activities in your community," a 49-49 split. Also the net seems to better serve national and international news as opposed to local news (see top of table). That makes sense, as the Internet destroys geography and cobbles us back together based on interests, partisanship, and the rest.

Those are my two takeaways. There's no a lot here otherwise, in part because it appears this is part of a bigger survey in which the results are not yet out.

We're All Good Drivers. Or so We Think

We all believe we're better than average drivers, according to a new survey.  What's wrong with this picture?

1. We're also mathematically challenged.  After all, we can't all be better than average given the very definition of average.
2. The world is far too full of bad drivers for the above statement to be true.
3. I, on the other hand, am better than the average driver.

People often overestimate themselves, particularly on how much they know.  Actual knowledge and perceived knowledge do correlate with one another.  In other words, survey a thousand people and ask them questions to measure what they know and what they think they know and the two tend to be positively correlated.  In general, those who tend to think they are more knowledgeable also happen to be more knowledgeable.  But there are exceptions.  It's where perceived knowledge outstrips actual knowledge that people get into trouble.

According to the story:

U.S. drivers believe their own driving knowledge, ability and safe driving habits are better than other drivers on the road, 64% rating themselves as “excellent” or “very good” while dishing out those high marks to only 22%t of other drivers their same age. Even among close friends only 29% of motorists ranked their buddies “excellent” or “very good” drivers.

We always overestimate (lie about) certain positive behaviors (voting, church attendance, gas mileage for our SUVs.  We tend to underestimate our negative behaviors (eating at McDonald's, watching Sponge Bob, picking our nose).  And when it comes to estimating the abilities of others?  Well, we tend to underestimate the positive, overestimate the negative.  Such is human nature.

So what's all this mean?  In part it's based in self esteem.  We want to feel good about our abilities and knowledge.  We need to.  I did not use "need" accidentally, because this is a major motivational factor in how we think about ourselves and our world.  We see the same thing when asking people what they know about politics and what they think they know about politics.  Certain kinds of news consumption (talk radio, for example) tends to increase our perceived knowledge but not really doing much for our actual knowledge.  TV news has the same effect, but for those of the lowest educational level or knowledge, TV actually does help.  This is a function of how TV tells stories and a function of those who tend to rely on TV and not print.

When it comes to driving, I'm not surprised by the result.  But the story itself is quite interesting.  Teen and senior drivers get rated the lowest, and that's no doubt true.  It's worth a read and worth thinking about outside and beyond driving as a domain. 

Wanna Learn? Take the Test

Taking a practice test beats studying -- even cramming -- when it comes to learning.

A NYTimes story on this research is all about efficient retrieval of information, a topic near and dear to the heart of this blog.  It's based on a pair of careful experiments that pit four approaches to learning.  Rather than spend the words here repeating the methodology, you can check out the links yourself.

But for me here's the really cool part:

In the experiments, the students were asked to predict how much they would remember a week after using one of the methods to learn the material. Those who took the test after reading the passage predicted they would remember less than the other students predicted — but the results were just the opposite. 

 I love this -- people who took the practice test perceived themselves as learning the least when, in fact, they did the best when it came to learning the material.  This gets into projection, into expectations, and probably a little bit of obsessive-compulsive behavior due to a lack of good, solid cramming.  When we take a practice test, we recognize gaps in what we know and at the subconscious level perhaps we take care of business, figuring it out.  There's a lot to be explored here.

This isn't easy to translate into the real world of media, politics, and knowledge -- I just find it fascinating -- but it does get at our ability to underestimate, and overestimate, what we know depending on the kind of "news" we consume.  I've written at length about how certain "news" or media content can lead more to the perception we are informed rather than actually informing us (entertainment programs such as The Daily Show, talk radio, etc.).  I'm certain this recent study is getting at the same or a similar mechanism.  And that's heartening.

What Men Think Women Know

I've written extensively about the relationship between what people know and what people think they know -- that is, actual knowledge versus perceived knowledge.  And, fascinated as I am with research methodology, I've also blogged about studies that consistently find men do better on tests of political knowledge than women (lots of reasons for this, some methodological, some cultural).

And now, a new study comes along that kinda combines the two.

Published in a recent Political Research Quarterly, the study finds that both men and women perceive women to be less knowledgeable about politics -- regardless of the actual knowledge each person has.

Here's an interesting bit, written in socialsciencespeak:

Though Huckfeldt (2001) asserts people can judge the expertise level of those with whom they discuss politics accurately, the perception effects we find here suggest otherwise; there is distortion within the judgment of expertise according to one’s gender. We suspect that even though discussion continues to take place at comparable levels between men and women main respondents and men or women discussants, the assessment that women know less about politics, even if that is not truly the case, may affect the dynamics of discussion within the dyad.

In other words, men assume women are not knowledgeable about politics without actually assessing what they know about politics.  The study goes on to discuss a number of potential consequences, all of them important.  A useful study if you're interested not only in political knowledge, but perceptions of knowledge and gender differences.

Perceived and Actual Knowledge

There's what people know, and what people think they know.  In this piece, Nicole Skibola in a Forbes blog notes:

In his book The Wisdom of Crowds, James Surowiecki discusses how disconnects between status and knowledge often lead to inefficient outcomes in decision-making processes. Groups, and obviously business organizations, tend to give deference to individuals with perceived or actual status, often with little to no correlation with actual knowledge.  As the group coalesces around one perceived knowledge holder, minorities are less likely to share their own perspectives, despite the fact that minority opinions make groups wiser as a whole.

Think of it as a micro version of the spiral of silence in which minority viewpoints tend to spiral into, yes, silence, because the majority speaks up so much, and so loudly, and our fear of isolation is such that we don't dare speak out.

I should point out that, despite what the book suggests, actual and perceived knowledge are typically correlated with one another.  Not perfectly, mind you, but there is a tendency for those who know something to also believe they know something.  I note the imperfection of the relationship because I've been reading a study on misinformation and how confident people are in their completely irrational beliefs, such as whether Barack Obama is Muslim.  The more misinformed, some studies suggest, the more confident people are in that misinformed belief.  In other words, it's damn hard to shake someone loose from crazy beliefs, even with facts.

Being uninformed, and being misinformed, are two conceptually distinct matters.  More on this another time.

What People Think They Know

As interesting as actual knowledge can be as a topic of research (or a blog), even more interesting are perceptions of knowledge -- not only what people know, but what people think they know.  In this case, let's talk global warming/climate change.

The fine folks at Gallup have asked: "Next, thinking about the issue of global warming, how well do you feel you understand this issue -- would you say very well, fairly well, not very well, or not at all?"  The bold face emphasis is mine.  How much do they think they know.  The results?

From 2006 to 2010, the numbers remain largely unchanged, but there is a tick upward in 2010 compared to previous years in what people think they know. See below:

              Understand

              Very Well

2006            21%

2007            22%

2008            21%

2010            26%

Now you might find this bump heartening, but I suggest it's actually a bad thing.  Let me explain.  We've seen a drop in the proportion of people with a clue about science and who doubt global climate change -- sometimes for good reasons, sometimes because for silly partisan ones.  But I suspect this has bumped up the perception that people are knowledgeable when, more likely, they confuse strong feelings about a topic with knowledge about that topic.

I know of what I speak. I've done many studies looking at actual knowledge and how it compares with perceived knowledge.  The two are correlated, but imperfectly.  And often emotion gets in the way and makes people feel informed when, instead, they only think of themselves as informed.  While I don't have data here to explore this further, my educated hunch is this bump in 2010 of perceived knowledge about global climate change masks the role of ideological silliness, partisan pettiness, and a sense of being informed through emotion than really being informed. 

It'd be a fascinating topic.  Hmmmm.  Maybe time to go there.  Off to search for good, solid data.

Objective and Perceived Knowledge

I've written many times about perceived knowledge, the perception that we are informed, and how it is different from -- but related to -- actual or objective knowledge. Indeed the two are often highly correlated, but not perfectly so. My interest has always been in the role the media play in creating the sense of being informed versus actually gaining knowledge.

But let's talk condoms.

There is this study, published in Pediatrics, that examines perceived and objective knowledge about adolescent male condom use. Other than being a neat topic, I found the methodology kinda interesting.

No it's not what you think, so keep reading.

They used a 5-item scale to measure knowledge and a 5-item scale to measure confidence about that knowledge. Is confidence the same as perceived knowledge? That's an interesting question; it's fascinating to tap not only a set of knowledge items but confidence about those very same items. So if you missed a few of the five but were very confident, you are high in perceived knowledge without being high in actual knowledge. Typically we ask a very different kind of question to measure "perceived knowledge," a set that often look very much like internal efficacy. I think the approach here may be much better, given that it's domain specific rather than general, like efficacy.

Why does all of this matter? Simple. I argue that some news programming, especially entertainment-based news programming, can lead to the sense of feeling informed (my empty calorie theory) when you're actually not all that informed, or at least not as much as you believe. Feeling informed, those empty calories, mean you don't eat your spinach (serious news, that sort of thing).

In the abstract of the condom study, the authors report:

However, those with higher perceived knowledge, particularly in the context of low objective knowledge, may be at greater risk for not using condoms. Addressing not only objective but also perceived knowledge may increase the effectiveness of interventions that are designed to increase rates of condom use among male adolescents.

In other words, high perceived knowledge but less actual knowledge is a bad combination. I'd make the same argument across a number of domains, from health to consumption of serious news as part of being a citizen in a democratic society.

Actual and Perceived Knowledge

Interest in the difference between what people know (actual knowledge) and what people think they know (perceived knowledge) crosses a number of scholarly disciplines. I've done work on whether listening to talk radio or TV faux news shows increases knowledge or merely the perception that one is informed.

I'm not alone in my fascination with the difference between actual and perceived knowledge.

For example, this study finds elementary school teachers often overestimate what they know about literacy development in children. There's similar work in pharmacy, counseling, and political science (the latter gets into some fascinating gender differences and I hope to write more about this tomorrow).

My point? What people think they know can have a number of consequences, thus should be studied when looking at any type of standard knowledge. What consequences? If a person feels informed, he or she may engage in early closure, not seek out additional information, perhaps reach a faulty conclusion. Persons may fill up on "empty calories" of junk news or entertainment-based news and think themselves informed when perhaps they're not truly informed.

Perceived knowledge can lead to greater feelings of efficacy. That's the good. But if those feelings of efficacy are based on weak, insubstantial actual knowledge, you have to worry about any final judgments -- be it in the classroom with teachers, or in the voting booth by citizens. Or in any number of possible scenarios, from doctors to government workers.

I've always wanted to dig deeper into the theoretical possibilities of actual versus perceived knowledge, perhaps even write a good NSF grant aimed at what the consequences of this are for democracy. Maybe I will. Some day. But I firmly believe that if you're studying knowledge, you need to also study perceived knowledge.