Editorial Standards

These are the working rules behind every page here. They are printed for one reason: so that a reader can take any article on this site, measure it against the list, and catch it where it fell short.

Why this site exists at all

Search almost any question in this subject and count who published the results. The pages telling you whether your teenager needs treatment are published by people who sell treatment. The pages describing what a social skills programme involves are published by the people running the programme. The pages about whether this is shyness or something more are published by clinics that would charge to assess it.

None of that is lying. It is a room where nearly everyone answering has a stake in the answer, and that produces a reliable tilt. Every situation looks a little more serious than it is. Doing nothing is never one of the options on the list.

Nobody in that room can afford the sentence "you probably do not need to buy anything." This site can, so it says it wherever it is true. That is the whole reason to build it, and it is stated here once rather than repeated on every page, where it would just read as defensiveness.

Nothing prints unless it is already in the file

Facts on this site do not come from memory and they do not come from other people's articles.

There is one research file behind the site. Every entry in it is numbered, carries the exact wording of the source, and carries the link. A claim that has no entry does not go on a page, even when it is obviously true, and even when every competing page says it. That rule is inconvenient about twice a week and it is the only reason any of this is worth reading.

Three things follow from it:

Figures arrive with their period attached. A number gets the years it was measured over written into the same sentence, in the prose, not in a footnote. A survey run in 2004 and a survey run in 2024 are different claims about the world. Blending them quietly is how a figure that was once right keeps circulating for twenty years after it stopped being right.

A missing number is reported as missing. Several of the things readers most want to know here have never been measured. How many friends a teenager needs. How long it takes to recover from losing a group. How many adolescents grow out of this without treatment. On each of those the honest page says nobody has published a defensible figure, which is more useful than a plausible invention and considerably harder to write.

Quoted regulations are quoted whole. No ellipsis inside a quoted rule anywhere on this site. A sister project once shipped a safety error for want of this. A regulation had been shortened in the research file, and every page built from it inherited the gap without ever seeing it.

The claims that are banned outright

The research file keeps a do-not-print list, currently seventeen entries. It exists because several of the most quotable lines in this subject fall apart the moment you open the paper behind them. A few of the entries, so you can see the standard:

• The tidy sequence of hours it supposedly takes to become someone's friend. That sequence is not in the study it gets attributed to. It comes from a university press release. The study's own tables give different numbers. The hours were estimates people made looking backwards. Friendships that had failed were left out. And no participant was a teenager.
• The friendship recession percentages. Real research, wrong population. Those figures describe US adults. The older half was also collected by telephone and the newer half online, which is a shift no honest comparison can ignore.
• The often quoted collapse in time with friends, from about 150 minutes a day to about 40. The age band runs up to 24, so it is mostly not teenagers. The series ends in a lockdown year. And it disagrees with the federal time-use survey by a factor of about two and a half. This site uses the federal series and says which years it covers.
• A number for how many friends anyone needs. No adolescent figure exists. Any page giving you one made it up.
• Any suggestion that residential placement is a recognised step for social anxiety. No guideline body, systematic review or government source recommending it for this could be found.
• A single length for the most heavily marketed teen social skills programme. Its operator gives two different answers on two of its own pages, the published manual gives a third, and providers advertise a fourth. The honest page reports the disagreement.

There is also a whole entry about how the best known prevalence figure gets used. When this site prints it, it prints it with everything attached: the National Institute of Mental Health estimates that 9.1 percent of adolescents had social anxiety disorder and 1.3 percent had severe impairment. That estimate is lifetime rather than current, so it describes who will meet the criteria at some point by 18. It uses the older DSM-IV criteria. It comes from a survey fielded between February 2001 and January 2004. Printed bare, with none of that attached, the number is technically sourced and practically misleading. On a page a frightened parent is reading at midnight, that is the worst combination there is.

Three things this site is not allowed to do

It does not diagnose, and it does not set up a conclusion for you to reach. No page here lists a set of behaviours in a way designed to make you decide something. No condition is offered as the explanation for anything. Where a diagnostic question genuinely cannot be avoided, the page describes the process of getting a proper evaluation and links out, and it stops there.

It does not promise confidentiality on somebody else's behalf. Where a service has published exact wording about what it does with what you tell it, that wording is quoted and attributed. Everywhere else, the site teaches you to find out for yourself rather than handing you a reassurance it has no standing to give.

It does not tell anyone to take a step they cannot walk back. Moving a teenager to another school, placing them in a residential programme, contacting another family about their child, escalating an ordinary social exclusion to a school. Each of those is described here in terms of what it involves and who can advise on it. None of them is recommended, in either direction, because the consequences turn on facts this site cannot see. For the same reason no page says what a school, a district, a court or a clinician will do. Practice differs by state, by district and by individual, and the pages say so.

What the date at the top of a page claims

It claims one thing only: that somebody sat down on that day and read the sources behind that page again. What it records is reading, not editing. Fixing a typo leaves it alone. So does reformatting. Strip that rule out and the date would be worth almost nothing.

Two parts of the site get re-read far more often than the rest. Crisis details, because a service withdrew one of its routing options in July 2025 and its own help pages still described it as available a year later, so an operator's own site is not proof of anything. And prices, because they move, and because publishing them is the most useful thing this site does.

Two components nobody types by hand

The crisis details and the price table are generated from the research file, not written into pages. A correction in one place propagates everywhere at the next build, and no page can drift into carrying its own slightly different version.

The build refuses to ship in three cases. If a page that ought to carry crisis details fails to render them. If any page names one of those services, in words or in digits, without them. Or if a text-message short code has been wired up as a telephone link.

Where a book may and may not appear

The person who writes this site, Jenna Hale, also wrote three books for teenagers. Eight of the pages here link to one of them. That is a financial interest, it sits in the footer of every page rather than only here, and you are entitled to weigh it.

The constraints on it are stricter than they look:

1. One book per page at most, and only the one that actually matches the page. Three titles on one page is a bookshop.
2. Always below the sources. Never in the top half of a page.
3. Never on a page carrying crisis details.
4. Never anywhere in the section written for teenagers.
5. Never on any page about paying for something. Not on the price pages, not on the page about whether therapy is warranted, not on the checklist for vetting a provider. Those pages exist to stop a reader being sold to, and that has to include being sold to by us.

Underneath all five: no page holds part of its answer back so that buying something looks necessary. A page that did that would be an advertisement wearing a reference page's clothes, and it would deserve everything that happened to it.

Corrections

Errors get repaired in the open. The page is fixed, the checked date moves, and where the repair changes what the page had been telling you, a line saying so stays on the page rather than being quietly tidied away.

If something here is wrong, send it over. A correction with the source attached is the most useful message this site gets.