From our Obsession
We’ve never been as connected, or as isolated.
Head Start, the childcare program designed more than half-a-century ago for low-income families in the US, has an unusual problem: It is both oversubscribed and under-attended.
Created in 1965 as part of president Lyndon Johnson’s War on Poverty, Head Start helps to prepare disadvantaged 3- and 4-year olds for formal schooling. A parallel initiative, called Early Head Start, also serves children under 3.
Overall, Head Start has fewer spots than the number of people who are eligible to fill them. Head Start programs serve more than a million (pdf, p. 2) infants and toddlers every year, but according to the National Head Start Association (NHSA), only 31% of eligible children between 3 and 5 years old and 7% of eligible children under 3, had access to Head Start services in 2017. And yet those who do manage to access Head Start don’t always take full advantage of it. “Somehow paradoxically,” says Ariel Kalil, a developmental psychologist at the University of Chicago and the lead author of a recent experiment conducted in Head Start centers, “they have a very large problem of chronic absenteeism.” Chronic absenteeism, meaning the percentage of students who miss 10% or more of school days for any reason, can be as high as 45% in some Head Start centers.
It’s not hard to see why this is a problem: If kids don’t show up to school, they have worse educational and life outcomes in the long run.
In the “Show Up to Grow Up intervention” (pdf), Kalil and her co-authors increased attendance and reduced chronic absences among Head Start families by sending parents personalized text messages three to four times a week to highlight the goal of having their child attend school every day. Though the intervention didn’t eliminate chronic absenteeism, it reduced it by 12 to 20%, adding to a body of work that shows something as simple as a few text messages every week can alter parents’ behavior and put their children on the path to success.
Why are some toddlers chronically absent from Head Start centers?
It really doesn’t take much to become “chronically absent” from school. A child has only to miss a couple of days a month to hit that 10% benchmark. Some Head Start programs only cover half-days, so if a parent is late leaving home, or there’s traffic, or the weather is really bad, it’s easy to see how they might just give up.
But chronic absences put a strain on Head Start centers and further handicap the very children who would most benefit from quality early childhood education. They can lead to sanctions for the centers, who are required to maintain an average rate of attendance of 85%. Chronic absences also deprive vulnerable children of crucial opportunities to learn, develop social and emotional skills, and get ready for formal schooling. And there’s evidence that once kids develop the habit of attending class at a young age, they are more likely to attend class later in life, so preschool could be laying the foundation for future attendance. As NHSA puts it, “a missed day of school is a missed opportunity for learning, and the implications are massive.”
“Absenteeism is a humongous problem,” says Kathy Hirsh-Pasek, a senior fellow at the Brookings Institution who sits on the advisory board of Vroom, an app developed by the Bezos Family Foundation to teach parents about babies’ “brain-building moments.” Hirsh-Pasek says, “It’s almost impossible to do a study anymore in a low-income environment where you don’t track the absenteeism as part of the outcome, as one explanatory variable.”
Head Start centers have come up with different solutions to tackle absenteeism, “mostly in ways that they tell us have not been particularly effective,” Kalil explains. That includes hiring a family engagement specialist whose job it is to shore up absences by calling families and asking them why their child is absent so often and what they could do to help.
There are both malleable and structural factors that can explain chronic absenteeism. Structural factors are those that have little to do with individual will or behavior and are difficult to change, like a family not having enough money for the bus fare, or a sick caregiver not being able to take their child to school. A texting intervention will do next to nothing about those. But then there are the malleable factors—the ones to do with habits, perceptions, and attitudes. They include the fact that preschool isn’t mandatory and many families don’t think it matters if their child misses a few days of preschool. That’s where texting can make a difference.
Using behavioral science to reduce absences
In “Show Up to Grow Up,” 741 parent-child pairs from nine Chicago Head Start centers were randomly sorted into two groups. Parents in the treatment group received three to four text messages a week for 18 weeks. The messages were personalized in that they included the child’s name and the number of days they’d been absent from school. Otherwise, they followed a similar pattern. One message focused on setting goals to attend school; another on prompting parents to plan around specific obstacles to attendance, such as transportation; a third on using data to correct parents’ beliefs about their child’s attendance; and a fourth on explaining what children would miss if they skipped school.
The researchers measured individual attendance rates, meaning the difference between attended days and enrolled days, using the centers’ records, and found that the average was 82% across all centers over the 18-week period. They found that 59% of the children were chronically absent.
During the 18-week intervention, children in the intervention group attended school on average 2.5 more days than their peers in the control group, an increase of about 4%. Children whose parents received the texts were 12% less likely to be chronically absent. When researchers defined chronic absenteeism more generously—as missing 15% of school days—the drop was 20%.
Then, the researchers compared these results with parents’ responses to a pre-intervention survey that asked 71% of participants about their views on the importance and value of preschool attendance. They found that the parents who rated preschool attendance as least important and who placed less value on preschool were the ones who most seemed to benefit from the intervention. This means that at least some share of chronic absenteeism can be explained by malleable, not structural, factors—and can therefore be changed.
“In this world, 20% makes a difference,” says Hirsh-Pasek. On the other hand, the intervention didn’t eradicate chronic absenteeism. “Clearly, for a number of parents, these messages didn’t do anything,” says Kalil. “Those parents may have gotten the messages and ignored them or thought they weren’t relevant for them.”
The study had other limitations. First, it didn’t shed any light on how the text messages reduced absenteeism. The texts, while personalized, were bundled into categories and “from a scientific perspective, it would be interesting to know which ones of those were driving the behavior change,” says Kalil. Second, there was no follow-up over time to see if attendance went back down after the text messages stopped. And there was no analysis of whether the effect of parents’ behavior change also spilled over to younger siblings who weren’t included in the study sample.
Kalil points out that, overall, parents did not drop out of the experiment, which they could have easily done by replying “STOP” to the text messages. But they have limited feedback from parents, because “we didn’t talk to [them] along the way,” she says. “That was part of the design, because to do so is expensive, and we wanted to see how inexpensively we could mount this intervention.”
While Kalil declined to specify the cost of the intervention, she said it was cheap by industry standards, and that the main expense was organizing preschool attendance records into a standardized file, because the quality of the record-keeping varied a lot by center.
The most important lesson of this intervention, says Rebecca Winthrop, co-director of the Center for Universal Education at the Brookings Institution, is that if you want to pull one lever to improve poor kids’ chances in life, your best bet are their parents. “Particularly for early childhood development, parent education and parent support is one of the most important types of interventions we can do,” she argues.
The experiment also adds to a body of evidence that text messages can nudge a change in human behavior, as set out by Richard Thaler and Cass Sunstein in their “nudge theory,” which describes the use of behavioral science to influence people in a desired way. Thaler and Sunstein, who wrote a book titled Nudge in 2008, described the tactic as follows: “Putting the fruit at eye level counts as a nudge. Banning junk food does not.” So, sending parents a text reminding them of the consequences of their child missing school days is a nudge.
Organizations like Vroom or Text4Baby, a free national initiative run by the National Healthy Mothers, Healthy Babies Coalition, already successfully help parents build their child’s cognitive and socio-emotional skills, or provide them with important maternal and child health facts. Experiments conducted in Kenya showed that weekly text messages to patients with HIV on antiretroviral therapy increased their chances of taking their medication every day. And an experiment conducted in clinics in the London Borough of Hillingdon found that sending women a text message two days before their first routine breast screening appointment significantly increased their chances of attending the appointment.
At their core, these initiatives are “just bringing the thing you’re supposed to do to the top of your mind so that you don’t forget to do it,” explains Kalil. ”We have to make [attendance] one of the top five priorities, instead of one of the bottom five priorities,” adds Hirsh-Pasek. “And if a text can move it up the priority chain, then we win.”
Correction: An earlier version of this post identified Ariel Kalil as a behavioral economist. She is a developmental psychologist.