Betsynote: Welcome to Part II of the interview with Andrew Flanagin and Miriam Metzger, two professors at the University of California Santa Barbara, who have just completed a first-of-its-kind study researching how adults and children are using online information. The prior section talked about who they studied, online behavior patterns as children aged, and the effects of the digital divide on the research. Part II focuses on how confident kids are, what cues they are using in the content to determine if it’s valid, what the research suggests should be done in teaching Critical Thinking in schools, and the road ahead for this kind of online research. The full results and findings from this study will be available via MIT Press this year.
Another thing that struck me was both the parents’ and the kids’ belief in themselves. Correct me if I’m wrong, but the study seemed a bit like Lake Wobegon, where everyone is good-looking and smarter than average — at least to themselves.
Yes, this struck us as well. The tendency for groups of people to believe themselves to be better than average in some way is pretty well-documented in social scientific research. In our study, we saw parents and children succumb to this bias too. Both groups saw themselves as better at figuring out good or bad information online, less likely to believe false information, and more likely to question information online than the average Internet user, although interestingly this effect was more pronounced in parents, who consistently rated themselves as better than their children across all three skills. Children agreed that their parents were better than they were at discerning credible information online, although older children thought they were less different from their parents than did younger children.
How did that self-confidence in Internet savvy strike you when you looked at the credibility assessments kids were making?
A majority of kids found at least some of the information on the Web to be believable. Similarly, a majority of the kids we asked reported thinking about whether they should believe information they find online at least some of the time, and thought others should do so as well. Kids were more likely to believe someone they knew in person than someone they met online, with almost half of kids reporting that they don’t trust people they meet online under any circumstances. However, a similar percentage of kids reported believing people they meet online at least sometimes. So it seems that kids are thinking about the credibility of information and people online pretty regularly, and doing pretty well really—although perhaps they’re not being quite as critical of people they meet online as they could be.
As more evidence that kids are fairly savvy in assessing the credibility of web information, our survey also revealed that in the context of online encyclopedias (for example, Wikipedia, Citizendium, and Encyclopedia Britannica) kids were more likely to believe information that was generated by “experts” from a well-established source (that is, information that came from Britannica rather than from the other two sources). They were also more likely to believe information that was balanced in its content than information that was one-sided in nature. Both of these findings seem to indicate that kids are pretty deliberate in evaluating both the source and the content of online information.
Another interesting finding that emerged when we asked kids to evaluate the credibility of online encyclopedias was that entries from Wikipedia that kids thought came from Encyclopedia Britannica were perceived as significantly more credible than entries about the same topic from Britannica itself. In other words, kids thought the actual content from Wikipedia was credible, but only when they thought it originated elsewhere.
And also, can you talk a little bit about how the genders and ages varied in terms of their Internet confidence?
It was interesting that even kids as young as 11 years old thought they were better than the average Internet user in terms of their search skill, technical skill, and knowledge about Internet trends and features. As the kids we asked got older, they believed themselves to be even more skilled or knowledgeable than the average Internet user – to a point. After age 15, kids’ average self-perceived skills no longer continued to rise, but leveled off or even decreased modestly. They still believed themselves to be more skilled or knowledgeable than the average Internet user, but the gap no longer increased as it had in previous years.
There were also intriguing sex differences with regard to kids’ self-perceptions of Internet skill. Overall, boys rated themselves as significantly higher than girls in terms of search skill and technical skill. When you factor in age, the differences between boys and girls become a bit more nuanced, but generally older boys saw themselves as more skillful than girls viewed themselves.
Content seems to play a part in whether kids find the information credible. Can you talk about what kinds of information kids found the most trustworthy and what kinds of cues they used to tell if the online information was credible?
Kids were most likely to believe information on the Internet about their schoolwork, followed by news information, then entertainment and health information (which kids were equally likely to believe), then commercial information, and finally information about people they meet online, which they were the least likely to believe.
It is obviously potentially problematic that kids believe entertainment and health information online equally, given that the consequences of believing false health information may vastly outweigh the consequences of believing false entertainment information. It is also unclear whether kids believe information on the Internet about their schoolwork the most because they process that information the most carefully, because they go to vetted or teacher-approved web sites, or simply because it’s convenient to do so given that they’re searching for information of that kind the most. However, it is encouraging to note that commercial information and information about people online are the two least believable types of information, as kids seem to recognize that those types of information may be presented with ulterior motives in mind.
We also asked kids about the credibility of news blogs and of Wikipedia, two sources of information that have emerged with the rise of the Internet. Although some kids were unsure of the credibility of news blogs or even what a blog was, the majority of those who had an opinion rated news blogs as significantly less credible than newspaper and television news. For Wikipedia, almost all the kids we asked had heard of it, and most had used it to look up information. Although not everyone understood the open-editing nature of Wikipedia, a majority of those that did said they found Wikipedia to be fairly believable, although interestingly they were slightly more skeptical about how much people in general should believe Wikipedia.
As for how kids figure out if information online was credible, we broke down their information evaluation strategies into three categories, based on previous research: heuristic (relying on their gut feelings, making decisions based on feelings, making quick decisions), analytic (carefully considering the information, double checking facts, gathering a lot of information, and considering all views), or social (getting advice from others or asking for others’ help). We found that kids were more likely to report using analytic strategies than either of the other two, although the tendency to report using all three types of strategies increased with age.
The most important cues or elements that kids used to determine information credibility online involved the currency of the information, the security of the web site, information completeness, and the authority of the information source (for example, if the information originated from experts). Next most important were a number of indicators of endorsement from others and reputation. Those cues reported as least important dealt with web site design and general feelings about the web site. Kids also differentiated among news, entertainment, health, commercial, and school-related information when choosing which cues to look at. For instance, kids tended to look at fewer credibility cues, and apply these cues with less rigor, when looking for entertainment information than they do when looking for news, health, commercial, or school-related information.
These findings are largely consistent with research on adults, except that web page design has been shown to be a particularly important credibility cue for adults for all types of information. The difference here could be due to kids’ greater tendency to answer the survey in socially desirable ways, or could stem from their heavy use of Internet information for schoolwork, where the stakes for using non-credible information are relatively high.
What about Internet vs. other kinds of media like TV, radio, or books?
In our survey we asked kids which medium (television, newspapers, Internet, books, magazines, radio) offers the most believable information. Answers varied depending on the type of information the kids needed. For example, the children in our study indicated that they trust the Internet more for health information than they trust television, newspapers, magazines, and radio. The Internet was also rated as the most credible source of information for schoolwork, and was rated highly (in relation to other media) for information about something they might want to buy and for entertainment information. Overall, kids appear to rely fairly heavily on the Internet for a variety of kinds of information and believe it is a trustworthy channel of communication, which is not too surprising for this generation of “digital natives.”
V. Teaching Critical Thinking
You guys also tested kids with “hoax sites” — can you talk a bit about how the children did in evaluating them and the effect of information literacy training on these scores?
We showed children one of two hoax sites (one on male pregnancy, one on a fake campaign to “Save the Rennets”). Overall, kids found these hoax sites to be highly unbelievable, with 48% saying they believed the information “not at all,” and 41% saying they believed it “a little bit” or “some.” Only 11% said they found “a lot” of it believable.
One explanation could be the high likelihood that kids in our study received media literacy training. 73% of kids reported having been instructed on how to assess the credibility of online information by someone like a teacher, librarian, or parent. Further, 84% of parents reported that they spoke with their kids at least occasionally about what information was believable online, as mentioned earlier. These findings further undermine the caricature of kids’ naiveté and gullibility when it comes to information on the web.
Knowing what you know now after years of research, what recommendations would you give to educators who want to teach Internet literacy and critical thinking?
There is both good news and bad news from our study, which suggest some educational strategies for parents and teachers.
The good news is that for the most part children appear to fairly critical consumers of online information, particularly as they grow older and gain experience online. Also, for the most part kids appear to not be fooled by overtly fake information online, they make some intuitive and appropriate distinctions among different types of information, and they apply a variety of strategies to discern good from bad information on the web. And, kids seem to benefit from literacy training and to learn from their own and others’ negative experiences with bad information online.
The bad news is that kids appear to be overly optimistic about their own ability to distinguish what is credible information online, they overestimate their own Internet and information search skills, they are occasionally fooled by intentionally misleading information, and they sometimes fail to apply the appropriate rigor to assessing particular types of information online. Again, these negative indicators are for the most part more pronounced for younger kids.
Obviously, kids are going to use the Internet, for a variety of tasks. The vast majority of children begin using the Internet early–between 2nd and 6th grades–with a majority of kids online by 3rd grade, and 97% of all kids online by the 8th grade. Given this level of use, and the good and bad news from our study, the question becomes how to make their online experiences as beneficial as possible, by teaching children how to work through issues of information credibility successfully.
Our findings suggest that it is important for educators and parents to engage kids in digital literacy lessons starting at a young age, and to find ways to signal to them that their literacy skills can always be improved. The most important factors we’ve found in developing a healthy concern about online credibility include accumulating Internet use over time, sharing both positive and negative Internet experiences, and early involvement and engagement of parents in children’s Internet use. In fact, because even older children still believe their parents are more highly skilled in discerning credibility than they are, parents may be a particularly important element in successful literacy training.
Do you have ideas about your next research project? Will you continue to look at credibility issues or has this study now suggested another area to dig into?
We are always working on multiple areas of interest. In addition to web credibility issues, I [Andrew] continue to look at how and why people share information online, various uses of social media, and the changing nature of collective action in an age of digital technologies. I [Miriam] am particularly interested in issues of trust in digital media environments, which includes my work on credibility online, as well as my ongoing research on how people navigate and negotiate their privacy in online contexts, particularly in social media environments.
As for work on credibility, it’s an exciting and important time. Talented researchers from multiple disciplines around the world are adding to the body of knowledge on the topic, and authors in the popular press are packaging this work to make it more accessible. In our own work, we’re increasingly interested in cognitive strategies for information processing and the social dynamics of information evaluation. Also, we’ve completed a survey of adults parallel to the kids’ survey, which will enable us to make some interesting lifespan comparisons and learn about web credibility among the entire US population. Anyone interested in keeping up with our work can find more information at: http://www.credibility.ucsb.edu/
Other posts of interest: