Conventional wisdom often suggests that such children are confused and so they take longer to talk. Or, parents may hear that any apparent delay is just an illusion because kids are little geniuses who can learn many languages quickly and easily.
"Both of these views are wrong," US psychologist Erika Hoff told the American Association for the Advancement of Science meeting in Vancouver this weekend.
"It is not the case that hearing two languages confuses the child and impairs their ability to acquire anything. But it is also not the case that children can magically acquire two languages as quickly as one."
Instead, psychologists should take a different approach to testing young children, one that measures their proficiency in both languages instead of just one.
When that is done, researchers typically find that the two tests add up to about the same level of proficiency as would be seen in a baby who is learning a single language.
"Children who are exposed to two languages... must hear less of each language than a child who hears only one and so it takes them longer to get the same amount of experience with each language," added Hoff, whose research has focused on highly educated bilingual Spanish-English families in south Florida.
Two kinds of tests have existed for decades -- the Language Development Survey and the MacArthur Bates Communicative Development Inventory -- in which parents answer questions about which words their child knows and how many word combinations the child has at around age two.
Since their inception decades ago, both paper-and-pencil surveys have been adapted into different languages, with as many as 20 variations of the LDS and more than 60 of the MacArthur Bates now out there, researchers said.
Even this low-tech approach has proven superior to modern methods, said Philip Dale, professor of speech and hearing sciences at the University of New Mexico, Albuquerque.
"Despite an understandable skepticism you might have about the ability of parents with limited training and a natural pride in their child, parent-report can be quite an accurate measure," said Dale.
Leslie Rescorla, professor of psychology at Bryn Mawr college in Pennsylvania, who devised the LDS in the 1980s, agreed that surveys can be very effective in identifying late-talkers by 24-30 months of age.
In the LDS, parents are given a 310 word checklist, and are asked to mark which words their child says. Average children have about 150 words at that age, and late talkers have 25-50.
Rescorla presented research on new versions of the LDS distributed in Greece, South Korea and the Netherlands, which showed similar results as have been seen in the United States.
For instance, eight percent of Greek children surveyed were found to be late-talkers, compared to nine percent of US children.
Knowing whether a child is a late-talker is important because it may point to disorders that could be helped with early intervention, such as autism, hearing loss, or mental impairment, said Nan Bernstein Ratner, professor of hearing and speech sciences at the University of Maryland.
"Late-talkers are at high risk for other developmental problems," said Ratner.
"If you have children who have problems with language and with reading, we have children who will not succeed in society."
Previous research has shown that as many as 20 percent of all children are late-talkers, but many of them are simply "late-bloomers" who catch up by age five, added Ratner.
"About four-fifths of children will recover. The problem is we don't know which ones, so if you don't have a crystal ball it is much better to catch them at age two and to start tracking them, rather than to wait and see what happens."
When it comes to bilingual babies, Hoff urged parents to take advantage of the second-language tests out there, rather than worry about the child's poor scores in a single survey which is missing the full extent of the child's knowledge.
"Because these inventories are available in multiple languages, you can often assess what they know in both languages," she said.
"When you do that, the bilingually developing children look exactly like the monolingually developing child. They are acquiring total language knowledge at the same rate."