You Shouldn’t Have to Guess What Your Teenager Is Learning

You ask how the project is going. Your teenager says, “Fine.”

That may be perfectly honest. It still leaves the two of you with almost nothing to talk about. A better conversation starts when something concrete is on the table: the first attempt, a test that didn’t go as expected, a note about why the approach changed, or a finished piece next to the draft it came from. The question is how to have that second kind of conversation without turning your home into a monitoring station.

Activity isn’t understanding

Most of what we can easily count about learning describes use: logins, hours, streaks, tasks marked complete, badges. These numbers are cheap to produce, so they tend to be the ones we’re shown. They’re poor stand-ins for learning.

One small piece of evidence comes from the Harvard physics trial in which students learned a topic from a custom AI tutor. Most AI-group students spent under an hour on the lesson, and the researchers found that, in that one course, time on task didn’t correlate with post-test scores. Minutes spent say little about what was learned.

Streaks and completion counts have a second problem. Once a number becomes the goal, people start optimizing the number. A student asked to log daily activity will log daily activity, and a student rewarded for streaks will protect the streak. That moves attention from the work to the record of the work.

There’s also an honest limit to any record. Much learning leaves no trace. A teenager can think hard about a problem on a long walk and come back with a better idea and nothing to show. “No evidence recorded” should be read as exactly that. It doesn’t mean “did nothing.”

The change in the work is the evidence

What a record should capture is what changed. In a 1958 Paris Review interview, Ernest Hemingway said he rewrote the ending of A Farewell to Arms, the last page, thirty-nine times before he was satisfied. Asked what had stumped him, he said: “Getting the words right.” Decades later his publisher reissued the novel with alternative endings drawn from his papers at the John F. Kennedy Presidential Library, forty-seven in all. His count and the archive’s differ; the point survives either. Anyone reading those drafts learns more about Hemingway the writer than the published page can say, because the drafts show what he rejected as well as what he kept.

A very different case comes from a widely used teaching video. Ron Berger of EL Education shows a group of young students the six drafts of a butterfly drawn by Austin, a first-grader in Boise, Idaho. The students offer critique on each draft, guided by a standard Berger asks of them: kind, specific and helpful. The finished butterfly is lovely, but the six pictures lined up together are what show how learning happened. A parent who saw only the last one would admire it. A parent who saw all six could ask Austin what he noticed.

Now imagine this at teenage scale. A fifteen-year-old named Imani records a demo of a song. The first take has a chorus that drags. After a friend who plays says the second verse loses the beat, she shortens it and rerecords. The third version moves the key down because she can’t hold the high note. Hearing the three takes in order tells a parent what she heard and what she decided, which no single recording would.

Four things a parent should be able to find out

Four questions cover most of it, and each should be answerable from actual work, not from counts:

• What’s your teenager working on, and why?
• What have they made, tried, decided or changed recently?
• Where is there a documented difficulty, and what comes next?
• If they work with a coach, what does that person actually observe?

Each answer should point back to something real: a version, a test result, a student’s own note, a coach’s written observation. An answer that’s not tied to a source, such as “shows strong curiosity this month,” should be treated with the suspicion it deserves.

It also helps to know who’s speaking. “She revised the chorus twice” is a fact. “She seems more confident” is someone’s interpretation. Blended together, they produce confident verdicts about grit or character that nobody actually observed.

Visibility without surveillance

The risk in everything above is obvious. A family can turn “seeing the work” into watching the teenager.

Information, used well, helps. In 22 middle and high schools in Kanawha County, West Virginia, Peter Bergman and Eric Chan sent parents weekly text alerts about missed assignments and attendance, plus alerts about low grades. Course failures fell 27 percent and attendance rose 12 percent; test scores didn’t change significantly. Specific information gave parents something specific to do.

How parents come to know what their teenagers are doing points somewhere less obvious than more monitoring. In a study of 1,186 Swedish fourteen-year-olds, Margaret Kerr and Håkan Stattin found that what teenagers chose to tell their parents did more to explain the benefits of parental knowledge than parents’ tracking and surveillance did. A 2009 study by Eaton and colleagues reported the same pattern: parental knowledge correlated more strongly with child disclosure than with parental questioning. These studies are correlational, but the implication is practical. Teenagers who feel safe talking tend to be known by their parents, and there’s reason to worry that teenagers who feel watched talk less.

School monitoring points the same way. A 2022 report from the Center for Democracy and Technology found that about half of students say they’re not comfortable expressing their true thoughts and feelings online when they know they’re being monitored, and that most teachers at monitored schools say it exists to catch rule-breaking. That’s school software, not family practice, so the parallel is imperfect, but it shows the cost of record-keeping used to police.

Put the studies together and the useful kind of knowing is specific and about the work, and it can be had without the teenager feeling watched. One more rule comes from a value rather than the research. The alert study didn’t ask whether students knew what their parents were told. Even so, a teenager should be able to see whatever their parent sees about their learning, including any interpretation of it. A shared view turns a record into something to talk about, instead of a second, secret assessment. It also keeps the work in its owner’s hands: a parent who can read the draft is a partner, and a parent who rewrites it has taken over.

Some things make any record worse: rankings against other teenagers, and labels like “behind” on a project that’s simply stuck. Difficulty is ordinary learning information, and a teenager who stalls on a hard part is usually learning something about the part.

“Shouldn’t teenagers have room to experiment without constant scrutiny?”

Yes, and the argument above depends on it. Not every draft needs an audience, and a notebook can be a private place to be wrong. The line is between ordinary learning information that both the teenager and a parent can read and understand, and a punitive monitoring regime. A parent who’s responsible for and invested in a teenager’s education has a legitimate interest in a truthful picture. A teenager has an equally legitimate interest in knowing what that picture says. When there’s nothing recorded for a stretch, the right response is curiosity.

A better question for the car ride

Instead of “Did you finish?”, try: Show me one thing you tried and how it changed. If your teenager has something, you have a conversation. If there’s nothing, ask what’s going on without assuming the worst. The first attempt doesn’t have to be good. It only has to exist, and you have to be interested in it.