Say you’re unsure whether to trust something online. You ask someone who moves quickly through apps. They find the page at once. But you still don’t know why its claim deserves your trust.
Finding the page and judging its evidence are different parts of the problem. Calling someone “good with technology” hides that difference. A more useful question is what they can do or explain for this task.
This module is for adults who are learning or judging an everyday digital task. It pays attention to each person’s access needs and to what each study actually measured.
What you’ll get from this module
In this module
By the end of this module, you’ll have a clearer way to choose digital learning or help. We’ll connect studies of practical tasks, digital knowledge, older adults’ learning and deceptive emails at work. You’ll choose an everyday situation. Then you’ll use the exercise to pick out a skill to build or to get help with.
Ask Jing
Become a Jingius.
What does being good with technology tell you?
Performance study: operating and finding useful information.
Alexander J. A. M. van Deursen and Jan A. G. M. van Dijk’s 2011 study tested 88 Dutch-speaking adult Internet users. They worked on health-related tasks in a university office. On average, people completed 73% of eight operational tasks. That compares with 50% of three information tasks and 35% of two strategic tasks.
Operational tasks were basic actions, such as opening a website. Information tasks meant finding information. Strategic tasks meant using it to make a useful decision.
The percentages show how many tasks people completed in each group of tasks, averaged across participants. They don’t tell us what percentage of people were competent. The tasks also differed in difficulty.
The difference matters when you choose help. Someone may be able to operate a tool and still need help to find or use information.
Why it matters: Older participants tended to do worse at basic operations and navigation. For information and strategic tasks, the researchers found no statistically significant link between age and completing them in this sample. That doesn’t prove that ability is equal across age groups everywhere.
Read the Internet skills performance study.
Knowledge survey: age patterns depend on the question.
The Dutch study tested what people could do. A 2023 Pew Research Center report by Olivia Sidoti and Emily A. Vogels asked what people knew. It looked at nine digital knowledge questions put to 5,101 US adults. The responses were weighted to represent US adults.
Two-factor authentication uses two different checks to confirm who is signing in. Among adults under 30, 68% picked out an example of it. Among adults aged 65 or older, 26% did.
Another question asked how large language models work. These are the AI systems behind text chatbots. Here, 45% of adults under 30 answered correctly, compared with 15% of those aged 65 or older.
Why it matters: The survey measured knowledge. It didn’t measure how people behaved when they faced a security risk. The age differences in these answers were real. Dismissing them altogether would miss part of the evidence. But a group average can’t tell you what your own helper understands.
Read the digital knowledge survey.
Learning study: older adults improved headline judgments.
A gap in someone’s knowledge also raises a question: can they learn what they need? Ryan C. Moore and Jeffrey T. Hancock’s 2022 study looked at MediaWise for Seniors during the 2020 US election. It compared 143 volunteers who completed the course with 238 people who did not take it. Those people were recruited separately.
The online course took roughly one hour. It covered several ways to evaluate information. Participants judged whether news headlines were true or false.
The researchers used a statistical model to estimate the chance of judging a headline correctly. They left out neutral, undecided ratings. Among course participants, that estimated chance rose from 64% to 85%. In the comparison group, it went from 55% to 57%.
Participants chose to take the course. The researchers did not assign them to it at random. The two groups already differed before the course began.
The follow-up happened straight away, so the study doesn’t tell us how long the gain lasted. It also couldn’t show which part of the course produced the gain.
Why it matters: These older adults did improve on the headline task. Their starting difficulty wasn’t the end of the story. That is a reason to look for learning that fits a specific gap. It is not a reason to use birth year to decide who can learn.
Read the older-adult media literacy study.
Workplace experiment: support at the decision point.
Learning is one possible source of help. Support during a hard decision is another. Daniele Lain, Kari Kostiainen and Srdjan Čapkun’s 2022 paper looked at simulated phishing at one company. Phishing uses deceptive messages to trick people into unsafe actions.
Their author version describes 14,733 employees who received eight simulated phishing emails over fifteen months. Employees were assigned at random to different warning and training conditions.
Warnings reduced dangerous actions in these simulations. One such action was entering account details into a deceptive website.
A voluntary training webpage was shown after a failure. It didn’t help employees resist later phishing emails. It could make them more likely to fall for them. That finding is about this way of delivering training. It isn’t about all security education.
The age pattern didn’t fit a simple ranking either. Employees aged 18–19 were more likely than every other age group to click the phishing links and take dangerous actions. Those aged 50–59 were more susceptible than those aged 20–29 and those aged 60 or older.
Why it matters: The researchers saw these age differences within one company. They didn’t test whether age caused them. The result gives you a reason to look at the security decision in front of you and the support you can get. It is not a reason to treat youth or experience as protection.
Read the workplace phishing study’s author version.
What this means for you
Choose help for the part of the task you need to understand. Someone’s speed on a familiar interface may be useful. It can still leave your question about evidence or data unanswered.
Ask what the person understands about the question you have. Knowing how to use a tool is one thing. Explaining what happens inside it is another.
Separate the questions before choosing the helper
Start by separating four kinds of question. You only need to explore the ones that matter to your task.
Operating the tool: Can you complete the action and recover from an ordinary mistake? First check that you can read and use the interface. Finding it hard to use doesn’t settle how well you can judge information.
Evaluating a source: Why does a claim deserve your trust? Look for evidence beyond the page that makes the claim. Finding the page only helps if you can also judge what it says.
Judging a suspicious request: What makes the request uncertain or suspicious? Find someone with relevant security knowledge who can help before you rely on it.
Understanding the system and its data: What information goes into it, and who can receive that information? What does its answer or other output represent? These questions help you see what you still need explained.
Don’t fill a gap with a guess. If authoritative documentation doesn’t settle how a system handles your data, write down what is still unknown.
In the opening example, finding the page quickly answers an operational question. Explaining the evidence deals with your real concern. Ask an adviser to demonstrate or explain the ability that matters here. Their age isn’t the demonstration.
Match learning and support to the gap
If operating the tool is hard, work out which action needs help. Check that the interface is usable for you before you treat slow operation as a lack of judgment. Readable presentation, an input method you can use and time to understand the task all matter to this choice.
If evaluating information is the gap, look for learning focused on that skill. The MediaWise result supports the idea that people can get better at a specific judgment task. It doesn’t give you a reason to assume every course produces lasting gains.
For an unfamiliar or uncertain request, use any warnings and reporting support you have, and ask someone with relevant security knowledge. A reporting route needs people who can assess reports and act on them. Find out who that help is. Don’t assume that sending a report settles the question.
Treat uncertainty as a useful result
You don’t need to master every tool to work out your next step. Choose one gap that matters. You might need to learn an action, find someone who can explain it, or get information that neither of you has.
If the decision matters, try a different example later to see what you can now explain or do. Keep the task accessible and use a harmless familiar action for any demonstration. Never stage deceptive attacks or arrange success and failure by age.
Describe what you can now do or explain in that example. One success doesn’t settle every other digital task.
How about your daily life?
Where do you rely on someone because they seem good with technology? You might need help using an unfamiliar tool, checking a claim or understanding a request for information.
Write down a situation where you’d like clearer help. One suitable idea is enough.
| Situation | What I want to understand or do |
|---|---|
| ____________________ | ____________________ |
| ____________________ | ____________________ |
Which situation would be most useful to explore now?
My chosen situation: ____________________
Carry this situation into the exercise below. You can change your choice.
Identify your next learning need
Start with the everyday digital task you chose. If you came straight here, choose something you want to do or judge, and name the part that feels uncertain.
Keep any demonstration accessible and limited to a harmless familiar action. Never stage deceptive attacks or arrange success and failure by age.
- Describe the task. Write what you need to do or decide. Include any access needs.
Task and access needs: ____________________ - Separate its demands. Answer the questions below that matter to your task. “I don’t know” is a useful answer.
Operation: Can I complete the action and recover from an ordinary mistake? ____________________
Source evaluation: Why does the claim deserve trust, and where is evidence beyond the page making it? ____________________
Security judgment: What makes the request uncertain or suspicious, and who can help before I rely on it? ____________________
System and data: What information goes in, and who can receive it? What does the output represent? What remains unverified? ____________________ - Choose one gap and suitable help. Look for learning, or an adviser, that can deal with that specific part. Ask for an explanation or a demonstration. Don’t choose by age or speed.
For uncertain requests, use any warning and reporting support you have. Check that someone can assess a report and act on it. If authoritative documentation doesn’t settle how your data is handled, leave that question open.
Gap and next source of help: ____________________ - Plan a later check if the decision matters. Choose a different example so you can come back to what you can explain or do. If you can’t make that check yet, write down the missing information.
Later example or unresolved question: ____________________
Ask Jing
Become a Jingius.
Your next step
Finish with one clearly named learning need, source of help or unresolved question. Keep the result tied to this task. Don’t turn it into a general label for yourself or someone else.
A note on the evidence
The studies measure different things: task completion, knowledge answers, headline judgments and responses to simulated phishing. The people they studied and their settings also differ. Use those differences when you choose help. One result can’t stand in for every ability.
The headline study measured improvement straight after the course. The workplace study found that one way of delivering training failed. Specific learning and support deserve attention. But don’t assume every intervention gives the same result.
How we prepared this module
This module connects research on digital performance, knowledge, learning and security decisions. A made-up example about choosing advice explains the difference between finding a page and judging it. The written exercise helps you pick out one learning need or source of help for a specific task.
Sources
- Alexander J. A. M. van Deursen and Jan A. G. M. van Dijk. (29 April 2011). Internet Skills Performance Tests: Are People Ready for eHealth?. Journal of Medical Internet Research, 13(2), e35. Published article; Methods—Subjects, Procedure and Assignments; Results, Table 3 and Tables 5–8; Discussion—Principal Results.
- Olivia Sidoti and Emily A. Vogels. (17 August 2023). What Americans Know About AI, Cybersecurity and Big Tech. Pew Research Center. Original survey report; “How we did this” and “How Americans’ digital knowledge varies by age.”
- Ryan C. Moore and Jeffrey T. Hancock. (9 April 2022). A digital media literacy intervention for older adults improves resilience to fake news. Scientific Reports, 12, article 6008. Published article; Methods, p.3; Results—News headline classification task, p.4; Discussion limitations, pp.7–8.
- Daniele Lain, Kari Kostiainen and Srdjan Čapkun. (2022). Phishing in Organizations: Findings from a Large-Scale and Long-Term Study. IEEE Symposium on Security and Privacy, 842–859. Author version, arXiv submission 14 December 2021; III-C–III-E, IV age comparisons and VI-A–VI-B.
