Will my child be fluent as an adult? What is known

Nobody can promise it. The sources behind this app give no figure for how likely a child is to be fluent at 18, so this article gives none. What those sources do support is narrower: children and families differ a great deal, and the amount and quality of real interaction in the language is something a family can influence.

Why there is no number here

We looked for one. King and Mackey, in The Bilingual Edge, discuss individual differences between children, including birth order, gender, personality, aptitude and learning style (ch. 5). On gender, they conclude that each child is an individual. Gass and Selinker, in Second Language Acquisition, call heritage learners a heterogeneous group whose input varies widely (section 2.3). Our reading is that a group this varied cannot be summed up by one percentage.

Any figure offered as the answer to this question would be invented, so we do not print one. If you have seen a confident number elsewhere, ask where it comes from.

Numbers you may have seen, and what they are

Two findings circulate in this area. Both describe other families and neither predicts what will happen with your child.

• Birth order. As King and Mackey report it, a survey of 204 Korean-American families by Shin found that 79% of first-born children spoke Korean with their parents, against 66% of second-born and 43% of third-born. We read the book's account, not the study, and the book gives the study different years in different chapters. It describes those families at one point in time. This is REPORTED.
• A 20% line. As King and Mackey report it, Pearson and others (1997) studied 25 Spanish-English children and found that as little as 20% exposure was enough for a spoken vocabulary, while below about 20% children were very hesitant to speak. This concerns children's spoken vocabulary, not adult fluency, and it is debated. This is CONTESTED and we leave it unresolved.

What does fluent mean for your family?

Fluent can mean chatting with a grandparent, reading a novel, or studying in the language at school. The plan generator asks for one main goal out of four: everyday speaking, reading and writing, connection with relatives, or support for a school program. The sources we read do not say which of these a child will reach by adulthood. Choosing the one that matters to your family turns an unanswerable question into one you can plan for.

What a family can influence

• Real interaction time. King and Mackey say bluntly that half an hour a week will not make a fluent speaker, however good the program (ch. 8). For the early years they suggest aiming at roughly 80% of home interaction in the heritage language (ch. 6). That target is for the early years only.
• A plan that fits the child's age. The plan generator changes its activity mix across four age bands, and the audit shows your real hours rather than your intended ones. The weights in that mix are the app's own design choices.
• A review when things change. A slip is information, and a revised plan is more useful than an abandoned one.

None of this is a guarantee. These are the things within reach, and they are worth doing because they are within reach, not because they promise a result.

What is still unknown

• The chance of adult fluency for a child with a given amount of exposure. No source we read gives it.
• Whether the early-years target holds through the school years. No source we read gives a later figure.
• Whether the birth-order and 20% findings hold for your language pair and your country. They describe the families in those studies.

A realistic goal for this year is smaller and more useful than a promise about age 18: pick the kind of fluency that matters to your family, count the hours you really have, and set the next month's plan to match.