Thirty million words.
For 20 years, a chasm of words has yawned between the children of college-educated professionals and those of high school dropouts, quantifying the academic disadvantage faced by the latter group long before they even start school. That statistic has led to a generation of vocabulary-centered interventions to close achievement gaps, including the federal Race to the Top Early Learning Challenge, the Clinton Foundation鈥檚 鈥淭oo Small to Fail鈥� initiative, and many others.
The 鈥�30 million-word鈥� gap is arguably the most famous but least significant part of a landmark study, Meaningful Differences in the Everyday Experiences of Young Children, by the late University of Kansas child psychologists Betty Hart and Todd R. Risley. As the work turns 20 this year, new research and more advanced measuring techniques have cast new light on long-overshadowed, and more nuanced, findings about exactly how adult interactions with infants and young children shape their early language development.
鈥淲e don鈥檛 want to just distill it down to a numbers game, because ... the important message to take away is not the poor versus wealthy families, but the opportunities children have to interact with rich language,鈥� said Dale Walker, an associate research professor in early language and communication and the director of the Juniper Gardens Children鈥檚 Project in Kansas City, Kan. She worked with Ms. Hart and Mr. Risley and continues the line of research.
鈥淚t鈥檚 not just throwing words at children, but making sure they hear new concepts, things of interest to them, so their brains make those connections earlier,鈥� she said.
By the Numbers
In Meaningful Differences, Ms. Hart and Mr. Risley tracked 42 infants just learning to talk, and their families, including 13 middle-class households, 10 each of professional and working-class backgrounds, and six living on public assistance. From the time the children were roughly 7 or 8 months to 3 years old, the researchers observed them for an hour a month, tracking how parents and children interacted and the children鈥檚 total word exposure.
The researchers found that, on average, children from professional families heard more than 2,150 words an hour. Those in working-class families heard about 1,250 words. Children in families on welfare heard little more than 600 words an hour.
鈥淚t鈥檚 not just the word gap; it鈥檚 what you use language for,鈥� said Barbara T. Bowman, a child-development professor and co-founder of the Chicago-based Erikson Institute.
Children of professionals also heard twice as many unique words, and twice as many 鈥渆ncouraging鈥� versus 鈥渄iscouraging鈥� conversations (鈥淲hat did you think of that?鈥� versus 鈥淒on鈥檛 touch that,鈥� for example.) By the end of the study, more than 85 percent of the vocabulary, conversational patterns, and language complexity of the 3-year-olds had come from their families, and children of professionals had vocabularies more than twice as large as peers in families receiving welfare.
Video: A Positive Parent-Child Interaction
Positive conversations between parents and children are about more than just how many words children hear, finds Dale Walker, an associate research professor at the Juniper Gardens Children鈥檚 Project in Kansas City, Kan. This video comes from a project stemming from the landmark 1995 Meaningful Differences study, and is meant to show the kinds of positive parent-child interactions that benefit children鈥檚 language development.
A follow-up led by Ms. Walker, using 29 of the children, showed vocabulary gaps in preschool predicted 3rd grade gaps in language-test performance. 鈥淲hat I found in visiting those children from kindergarten to 3rd grade was, those who had heard the least were still at a disadvantage years later,鈥� she said. 鈥淚 always knew where to find them; frequently, they were in the hallways, for behavior problems.鈥�
That doesn鈥檛 surprise W. Steve Barnett, the director of the National Institute for Early Education Research at Rutgers, the State University of New Jersey, in New Brunswick. Policymakers, he said, often acknowledge differences in exposure to words but not to 鈥渆ncouraging鈥� language.
鈥淚 think one of the least-appreciated implications of this [study] is the problem with how we segregate low-income children in preschool programs just for them,鈥� Mr. Barnett said. 鈥淐hildren were already replicating these [family] patterns in their own interactions. What did we think the consequences would be of kids who get together and interact with each other largely negatively?鈥�
The hourly estimates from those original 42 families were extrapolated to predict that by age 3, children of professionals would hear about 45 million words, compared with only 13 million for a child in poverty鈥攖he source of the 30 million-word gap. By age 3, a child鈥檚 IQ was more closely related to the number of words he had heard than to any other factor, including parents鈥� overall education or income level. A 2003 article in the American Federation of Teachers鈥� magazine American Educator solidified the best-known takeaway of the study in its title: 鈥淭he Early Catastrophe: The 30 million Word Gap by Age 3.鈥�
Meaningful Differences has spawned a generation of studies, including a network of more than 100 researchers in the United States and Canada. Subsequent research has borne out Ms. Hart and Mr. Risley鈥檚 conclusions, if not the extent of their extrapolation.
The Boulder, Colo.-based LENA Research Foundation, which uses wearable digital technology to record parent-child interactions for 16 hours at a time, calculates that children with an 鈥渆nriched language environment鈥� hear about 20,000 words a day鈥�22 million words by age 3鈥攚hile disadvantaged children hear half as many or fewer.
鈥淭here鈥檚 a lot of differentiation within the groups,鈥� said Jill Gilkerson, LENA鈥檚 director of child-language research.
鈥淎mong the low-[socioeconomic-status] families, it鈥檚 not the case that none of the adults in the house were engaging with their kids. There are some who are engaging a lot,鈥� she said. 鈥淎nd it鈥檚 also not the case that all high-SES parents are engaging a lot with their kids.鈥�
One LENA-based study found, for example, that the upper half of parents who had less than a high school diploma鈥攊n terms of parent-child verbal interactions鈥攕poke more to their children each day than the lower half of college-educated parents, 14,245 words versus 11,802 words.
Parent-Child Conversations
鈥淭his is the challenge of translating science to policy, and when one study captures the imagination of the public, and policy is made based on one study,鈥� Mr. Barnett said. A study 鈥渉as to be viewed in the context of the much larger body of knowledge about language and family and experience.鈥�
In this July 2014 video, John Tulenko of Learning Matters reports on Providence Talks鈥攖hen a pilot program鈥攖hat gets low-income parents in Providence, R.I., talking more to their toddlers.
SOURCE: Learning Matters
But if recent studies shrunk the word gap from the Hart and Risley study, they also magnified the importance of parent-child conversations.
鈥淐onversational turns are vastly more important than the number of words a child is exposed to,鈥� Ms. Gilkerson said.
In one recent study, Ms. Gilkerson and her colleagues found parents of children who scored in the top 10 percent on preschool language tests had conversations with their children that involved 18 more turns taken per hour than parents of children scoring in the bottom 80 percent.
鈥淲hen you鈥檙e getting more directive, business talk, the short directive sentences don鈥檛 elicit this back and forth,鈥� Ms. Gilkerson said. That finding could explain, in part, why more total words are associated with better language skills for children. 鈥淲e do know as the number of words increases,鈥� Ms. Gilkerson said, 鈥渟o does the richness, the syntactic complexity, the content.鈥�
Helping Families Communicate
鈥淚 think [Ms. Hart and Mr. Risley] came up with a very substantial finding and something we have to pay attention to,鈥� said Susan B. Neuman, the chairman of teaching and learning at the Steinhardt School of Culture, Education, and Human Development in New York City. 鈥淲hat I worry about is how it鈥檚 being interpreted. Just exhorting parents to talk more is a very simplistic response.鈥�
While many early-literacy interventions focus on getting parents to read to their children, Ms. Neuman found in a forthcoming study that low-income parents engaged in more and better conversations with their children while playing an animal bingo game than while reading a storybook.
鈥淟ots of times researchers want to bring to low-income families middle-class practices, rather than learning what really engages and interests those families,鈥� Ms. Neuman said.
In Providence, R.I., researchers are trying to listen. The city is scaling up a pilot of its $5 million Providence Talks initiative. The program uses home visits and audio recordings of families to help parents understand and improve conversations with their children.
Caitlin Molina, the program manager for Providence Talks, said families who entered with children performing in the bottom half on an early-literacy test boosted the number of average daily words by 65 percent, or 5,600, and the number of turns they took in conversations with their children by half, or 116 new interactions, as compared with a control group of children who were given reading materials but not recordings and feedback.
鈥淚 see [the improvement] as incredibly important,鈥� Ms. Molina said. 鈥淚鈥檝e worked with this age group for many years, and you can identify very early what their learning patterns will be. To be able to tap into that at an early age is a gift.鈥�
