Closing the ‘word gap’

Providence Talks hopes to close the “word gap” between rich and poor, reports NPR. The project is funded by a $5 million, three-year grant from Bloomberg Philanthropies.

Providence will distribute small recording devices — essentially word pedometers — that tuck into the vest of a child’s clothing. These will automatically record and calculate the number of words spoken and the number of times a parent and child quickly ask and answer each other’s questions.

The LENA Foundation in Colorado developed the device, which makes it easy to track verbal interactions. Parents will get coaching in how develop their children’s language skills.

The children of low-income, less-educated parents may be six months behind in language development by the age of two, a Stanford study estimates. By the time they start kindergarten at 5, they score more than two years behind.

The Providence experiment was inspired by Three Million Words in Chicago, another gap-closing effort.

Aneisha Newell says that program taught her to talk to her young daughter in new ways. She says she never realized bath time — with colors and shapes of bubbles and toys to describe — could be a teachable moment. She ended up breaking the program’s record for the most words spoken.

And then there was the moment her daughter — not yet 3 years old — used the word ‘ridiculous’ correctly.

University of Chicago Professor Dana Suskind, who started 30 Million Words, said sitting in front of the TV doesn’t develop language. It takes interaction between the caregiver and the child.

“We can’t just have people saying 30 million times ‘stop it!’ It’s got to be much more,” she says.

The parent should “tune in” to what the child is looking at, talk about it and ask questions that can create a sort of “serve and return” between parent and child.

In Chicago, adults and children spoke and interacted more after receiving feedback from the LENA recordings and home visits, reports The Atlantic. Suskind is advising Providence Talks.

“Close the word gap, advocates say, and you might close the achievement gap and maybe even disrupt the cycle of poverty,” concludes The Atlantic. 

I like the idea of working directly with parents rather than trying to create preschools to do — a few hours a day — what parents aren’t doing at home. I’d make videos modeling parent-child conversations — and throw in a little coaching on how to get a child to “use your words.”

About Joanne


  1. It is kind of sad that we have to spend millions of dollars to tell parents to pay attention to, and talk to, your children.

    • Maybe the message should include recommendations not to have kids until they are able and willing to pay attention to and talk to their kids.

  2. The gap is substantially genetic

  3. I wonder if these programs encourage parents to take their kids places and talk about the new things that the kids notice. As a parent, I don’t really enjoy having constant conversation at home – we talk at meals, when we read or play games, or when the kids need to talk about something, but I also tell my kids to go play – my sanity requires breaks from constant chatter. The parent magazines induce a lot of mom-guilt in moms who read them because they make it seem like a terrible thing to NOT be talking constantly to your kids.

    But, when we go on outings (museums, gardens, zoos, even the hardware store or grocery store) it’s easy to talk about what we’re seeing for hours at a time.

    It also helps to make use of a library – if my kids want to know about something that I don’t have a clue about, we get books. That being said, kids really do pick up on whatever the adults around them talk about. My kids had an interesting discussion over the benefits of rain in the winter, when the leaves had fallen and thus there wasn’t a lot of photosynthesis, over breakfast a few weeks ago. A friend, an anatomy professor, has children who refer to body parts by their latin names, because that’s what their mom does. Most parents know something about something (cooking, home repair, gardening, how to work a cash register, being a mechanic, music, sports), and they can teach that to their kids -there’s vocabulary associated with any topic.

  4. Thinly Veiled Anonymity says:

    One might imagine that the quality of conversation matters at least as much as the quantity. OK, so we tell parents not to say “Stop it!” 30 million times.

    But how do we close the “word gap” between a parent with a 6,500 word speaking vocabulary and a barely passable command of grammar — even when such a parent is convinced to engage at the very best of their ability — and a parent with a 22,000 word speaking vocabulary and an intuitive, fluent grasp of the logical structure of language?

    How do we — and forgive my bluntness here — turn the kids of poor- and working-class families into the kids of middle- and upper-class families? Because that’s really what I suspect all this “gap closing” is really about. And I’m not sure it can be done.

  5. This is all nonsense. Hearing children of deaf-mutes grow up to have statistically average IQ’s. How parents interact with their children within a broad range of normal behavior makes little difference.
    Thinly Veiled Anonymity – In most human languages a vocabulary of a few thousand words is what most speakers use.
    To lulu – You can speak to your children in classical Latin or Sanskrit. It won’t hurt them but it won’t make them any smarter.

    • Narrowing the word gap might not have an effect on IQ as an adult, but it would certainly help low income children be better prepared for school.

    • Jim, the effect of parenting differences on IQ is low for middle class families, but is is huge for low SES families.

      You are also wrong about the effect of speaking to children in a different language. There is ample evidence that being bilingual makes you smarter.

      EB, while we do not yet know if this program will produce differences in adult IQs, we do know from more recent adoption studies that the IQ of low SES children varies widely by parenting style, so there is reason to hope that the changes will be permanent.

      • cranberry says:

        I doubt being bilingual makes one smarter–at least if you want to cite the Italian study.

        From the study, “Fourteen bilinguals heard Italian and Slovenian, 2 heard Italian and Spanish, 2 heard Italian and English, 1 heard Italian and Arabic, and 1 heard Italian and Danish.”

        Gosh, in the city of Trieste, which infants are likely to be dual-language from birth? The children of academics, perhaps?

        In other words, comparing the children of academics to children of non-academics does not prove that “being bilingual makes you smarter.” If being bilingual were to make one smarter, I’d expect all the American-born English language learners in the US schools to be ahead of their peers.

        Being bilingual, but not ELL, in this country often correlates with well-educated parents who start the elite academic competition before birth.

  6. Lots of things correlate with lots of other things. For example, the following probably all correlate with each other:

    Vocabulary size of children;
    Vocabulary size of parents;
    Amount of interaction between parent and child;
    Amount of exposure to books;
    Amount of exposure to classical music;
    Age at which one begins reading;
    Number of books in the home;
    Measured IQ of child;
    Measured IQ of parents;
    Grades in school;
    Success in life;

    The big question is: what is cause and what is effect?

    • Florida resident says:


      • I don’t buy it. Low SES people aren’t low SES because of genes. If so, the vast majority of people would be serfs, since nearly all of us are descended from serfs of various kinds. Environment, nutrition (and toxins), all sorts of things play a role in how intelligence is expressed in a population. I think genes probably have less to do with it than many other factors, except that genes are way more complex than we thought anyway and get turned on and off in response to lots of things (again, like nutrition and environment).

  7. dangermom – You may have heard of the term “epigenetics” – You do not understand it.

    • Jim – You may have heard of the adoption studies comparing IQs of children adopted into high and low SES families. You do not understand these studies.

      • Richard Aubrey says:

        Apples and oranges. Poor people don’t get to adopt. Nor do single parents, usually, unless they’re gay in which case they can probably finesse it due to being in an Accredited Victim Group.
        So if a kid is adopted into a low SES family, it’s not a “poor” family of the type presumed when doing the word count work.
        A couple of parochial school teachers, one of whom is going to quit to take care of the kids, would qualify as pretty low income. SES means, iirc, Socio Economic Status. Socio and Economic are not welded together. In fact, the local Child Abductive Services might not think they’re qualified economically, but they would, if they managed to adopt, be pretty high up in the word count issue.
        So while there may be a difference–identical twins separated at birth and discoverable later by researchers not providing a huge sample group–the word count work is not looking at them so much as the really low SES.

        • Richard, it’s obvious you have no idea what these studies were about and what they concluded. Children adopted by low SES had much lower IQs than their sibling who were adopted by middle class families. This is not associated with the word count research.

          The rest of your rant is just weird and homophobic. Why don’t you try using logic and evidence instead of bigotry and emotion?

          • Ray. The Turkheimer paper doesn’t say anything about the effect of parenting differences it just attempts to assert that for low SES children environmental factors are more important than gentic factors. Whilst the paper is commonly cited many other investigations of low SES children have failed to find the same relationship. The only really large study 5000 pairs of twins, rather than 600, not only failed to replicate Turkheimer’s claims but suggested that it was probably statistically invalid to make claims based on twin studies under 4000 pairs. Handscombe et al PLOS one February 2012 | Volume 7 | Issue 2 | e30320. Moreover a recent paper established a clear link between Children’s IQ and SES which suggests that attempting to use SES as an independant variable in these studies is highly problematic as it is predominantly genetic not environmental. “Our analysis provides the first DNA-based evidence that the well documented association between family SES and children’s cognitive development, routinely interpreted as an environmen-tal effect, is substantially mediated by genetic factors”
            Maceij Trzaskowski et al Genetic influence on family socioeconomic status and children’s intelligence