By age and family shape
What works for a bilingual preschooler, ages 3 to 5
“What works for a preschooler?”
By the Kinlingua team · Updated October 19, 2026
The sources read for this site do not show that ages 3 to 5 decide how well a child ends up with two languages, so this article does not claim it. They do support habits that fit a preschooler: the heritage language as an everyday language at home, picture books read again and again, songs, and screens watched together. What they leave open is listed at the end.
What the sources say about the age
In The Bilingual Edge, King and Mackey recommend roughly 80% of child interaction in the weaker language for the early years, because the majority language seeps in from everywhere else. The methodology notes record no age range for the early years, so the sources read do not say whether the target covers ages 3 to 5 or ends sooner. They also do not compare ages 3 to 5 with any other age. The age bands in the app (0 to 2, 3 to 5, 6 to 9, 10 and over) are the app's own cutoffs, not taken from a source.
What to do at this age
King and Mackey give different advice by family type. This is their own advice, not a study:
- Both parents speak the heritage language: use it as the everyday language between at least one parent and the child, without worrying that this will hold back the majority language.
- Parents with different first languages: aim for roughly equal quantity and quality of exposure in both languages.
- Neither parent speaks it natively: do not worry about native-level accuracy, and weave the language into routines that already happen.
On top of that, these habits fit a preschooler, using the app's guidance for this age:
- Read picture books with a repeating phrase. The app suggests one or two sentences per page and a phrase the child can predict and say before you do. Re-reading is fine. The one reading study in these sources (Patterson, 2002, as King and Mackey describe it) involved two-year-olds: how often they were read to in a language predicted their vocabulary in it, even more than total exposure. It did not test ages 3 to 5.
- Sing songs with a story or counting. The app's content says this is the age when children start singing along, not only listening. That is app guidance with no cited source (UNVERIFIED).
- Watch screens together. The app's screen guide for this age, drawn from King and Mackey's chapter 7, says television and DVDs help only when watched together: pause to ask a simple question or repeat a line, and pick a word of the week from what you watched. The methodology notes do not record a passage on ages 3 to 5, so this is tagged UNVERIFIED below.
- Rephrase or offer a choice when your child looks confused. Gass and Selinker describe this repair pattern in conversations between non-native speakers. Those were not parent-child conversations.
- Do not make it a power struggle. King and Mackey write that you cannot make a child speak a certain language any more than you can make a strong-willed four-year-old wear a certain outfit.
What the plan generator gives a four-year-old
ILLUSTRATION from the app's real code, not data about real families. I ran the plan generator for a 4-year-old with Spanish at home, English in the community, one Spanish-speaking parent, 30 minutes a day and some existing exposure. With a conversational goal it returned 14 minutes of talk, 7 of reading, 5 of songs and 4 of a daily ritual. With a literacy goal it returned 12 of reading, 7 of talk, 4 of songs, 4 of ritual and 3 of a structured activity. Both add up to 30. For a 1-year-old with the conversational goal, the same code gave 17 minutes of talk, 8 of songs and 3 of reading, so reading gets more room as the child gets older.
The split by age is the app's design choice, with no cited source (UNVERIFIED), and so are the waking hours it assumes: 11.5 a day at this age, or 690 minutes. Thirty minutes is about 4.3% of that (30 divided by 690), a deliberate block and not the whole day. The app also sets a 30-day check, that your child starts a short exchange in the heritage language without being prompted. That is an app target, not a developmental norm.
What is still unknown
- Whether ages 3 to 5 matter more than other ages. No source read compares ages.
- How long the 80% target for the early years applies. The sources read give no end age.
- Whether the reading finding holds at ages 3 to 5. It studied two-year-olds.
- How much co-watched screens help at ages 3 to 5. The app's guide is not tied to a passage recorded in the methodology notes.