The federal Office of Management and Budget is increasing its push to get federal agencies to put their money where the research is.

In a memo to Education Secretary Arne Duncan and other agency leaders, the OMB has called for all fiscal 2014 budget proposals to include a separate section detailing the departments' "most innovative uses of evidence and evaluation."

Unlike the Program Assessment Rating Tool used to judge programs' budgetary worth—and often criticized for slow review cycles and findings of incomplete evidence—the memo calls for federal agencies to create and expand research partnerships to study programs, include cost-effectiveness calculations and embed the evaluation structure into program grants from the start.

In education, that could provide a big opening for the nation's regional educational laboratories, which in their most recent contracts were overhauled to include more partnering with state, local and other research groups.

The Education Department's Institute of Education Sciences, which oversees the labs, "wants us to develop research alliances and put people together around topics that are priorities for states," said Barbara Foorman, the first Commissioner of the National Center for Education Research and now head of the Regional Educational Laboratory for the Southeast Region. Foorman told me earlier this spring, as the lab got up and running, that in tight budget times, both the federal and state education departments are pressing for more research on programs: "They're looking for more evidence of effectiveness before they start writing checks."

The memo highlighted the Washington State Institute for Public Policy, which provides analyses of program effectiveness and return on investment for the legislature in that state.

OMB is also calling for agencies to look for quick, cheap and dirty approaches to the research itself. Traditionally, "gold-standard" randomized, controlled trials take five years or more and tens of thousands of dollars to conduct. To assist in cost-saving, the OMB has highlighted a brief by the Washington-based Coalition for Evidence-Based Policy suggesting ways to use the already-produced administrative data to create lower-cost, quicker-turnaround experimental evaluations.

"We strongly support this new effort," said Jon Baron, coalition president, in an email. "As demonstrated in fields such as medicine and welfare policy, such evidence-based approaches can greatly increase government's effectiveness in addressing critical national problems in social policy and other areas; and identify important opportunities for budget savings to help address the long-term deficit problem."

Moreover, the OMB memo hints that the Education Department won't be alone in using waivers to encourage states and districts to test out new approaches.

"One of the best ways to learn about a program is to test variations and subject them to evaluation, using some element of random assignment or a scientifically controlled design," the memo says. "OMB invites agencies to explain how they will use existing waiver authorities to evaluate different approaches to improving outcomes."

It also encouraged federal agencies to follow the lead of the labor and justice departments' "pay for success" models, in which private groups "invest" in promising interventions to solve social problems and are repaid by federal grants in return for showing progress.

Posted by Sarah D. Sparks at 8:34 AM | Permalink | Leave a comment | Recommend

It seems common sense that if student achievement is the measure of school improvement, the students themselves need to be engaged, yet it seems student motivation isn't often directly addressed in programs to improve student scores.

A new Center on Education Policy report argues that educators and policymakers often overlook the importance of student buy-in and motivation when planning school improvement initiatives. While no one system or incentive will encourage all students, CEP researchers argue that educators should consider what we know about student motivation when designing programs for school improvement.

The report from the Washington-based think tank describes four foundations of student motivation:
• Competence, in which students think they have the ability to do what is being asked;
• Control, or students believing they have choice in what to do and can affect the outcome;
• Interest, in which students perceive value in the task or learning; and
• Relatedness, or believing that doing the task or gaining the knowledge will gain them social approval.

For example, efforts to "personalize" the class environment—creating smaller schools or keeping teachers with the same class for several grades, for example—may improve students' social relatedness. Moreover, as previously reported, CEP highlighted that poorly targeted incentives do little good.

"Rewarding specific actions that students can control, such as completing homework, yields better results than rewarding accomplishments that may seem beyond their reach or out of their control, such as whether they earn an A grade," the report says.

Posted by Sarah D. Sparks at 8:20 AM | Permalink | Leave a comment | Recommend

While the Education Department and policymakers work to improve a list of "dropout factory" schools, a new study suggests there may be an equally problematic list of "drop-in" factories, schools in which a significant percentage of students attend sporadically.

The report, released this morning by the Everyone Graduates Center at the Baltimore-based Johns Hopkins University and the Get Schooled Initiative, sponsored by the Seattle-based Bill & Melinda Gates Foundation, found that, among the six states studied, chronic absenteeism ranged from a low of 6 percent in Nebraska (in 2010-11) to a high of nearly one in four students in Oregon (based on 2009-10 data.) In high-poverty areas, chronic absenteeism can go as high as one-quarter of all students in rural districts and one-third of all students in urban districts, according to report authors Robert Balfanz and Vaughn Byrnes of Johns Hopkins University's Center for the Social Organization of Schools in Baltimore.

States greatly differ on how to count unacceptable absences, but in the end, it doesn't much matter why a student misses school. Chronic absenteeism—missing 10 percent of the possible school days in a year—causes a sharp decline in academic achievement. Based on data from Georgia, Florida, Maryland, Nebraska, Oregon and Rhode Island—the only states that actually track chronic absenteeism— the researchers estimate one in 10 American students, or 5 million to 7.5 million students nationwide, misses a month or more of school every year. And as I've reported before, absenteeism isn't just a problem with older students; more than 10 percent of kindergartners nationally and more than 20 percent of Oregon kindergartners are chronically absent.

As in Mr. Balfanz's previous research on "dropout factories," he found chronic absenteeism is often concentrated in high-poverty schools. For example, the researchers found 61 schools in Maryland in which 250 or more students missed at least a month of school each year. If one of these schools happens to be identified as "persistently low-performing" for federal accountability—not an unlikely bet—its efforts to turn around may be hamstrung by having students who miss as much as six months of school during a five-year period.

Take Providence, for example. In 2011, the district was home to 19 of the 29 persistently lowest-achieving schools in Rhode Island. In the 2010-11 school year, the district had a rate of chronic absenteeism at 34 percent, or 8,000 students.

Chronic absences have been shown to increase a young student's risk of being held back and an older student's risk of dropping out, but the report found the effects go beyond school. In Rhode Island, researchers found more than 75 percent of all students involved in the juvenile justice system had been chronically absent.

"The existing evidence could not be clearer," the researchers conclude:

"Academic achievement from kindergarten forward, high school graduation, and postsecondary enrollment are all highly sensitive to absenteeism. Missing even some school can have negative impacts, especially for students who live in or near poverty. Missing a lot of school, at any time, throws students completely off track to educational success."

Posted by Sarah D. Sparks at 6:18 AM | Permalink | Leave a comment | Recommend

If you haven't checked out my colleague Christina Samuels' story on the data problems with the U.S. News and World Report's annual rankings of the best high schools in the country, you really should. Errors in state data reported to the federal Education Department led to inaccurate rankings for three of the U.S. News "best high schools," and led NCES to conduct a review of all 5,000 schools on that list.

Maintaining data quality in an ever-tighter economic climate is likely to be a hot topic at the upcoming National Forum on Education Statistics in July. At a time when education officials are increasingly dependent on complex student data systems to make policy and instructional decisions, it's vital that the data are accurate and timely. Yet in the midst of a budget crunch, support for the data systems has been under fire in some states, and most are still working to train the districts in how to correctly collect, report and later use information on their students.

Moreover, states turn over that flood of student information to the federal Common Core of Data, the giant national information warehouse run by the National Center for Education Statistics. The NCES received $108.7 million in fiscal year 2012 for managing all those school data, up about $450,000 from fiscal 2010 but still miniscule in comparison to the $68.12 billion overall discretionary budget for education in fiscal 2011. (President Obama has asked for $114.7 million for the NCES in fiscal 2012.) Of that, NCES Chairman Jack Buckley tells me about $5 million each year goes to the Common Core of Data, though other parts of the Education Department also contribute to data collection costs for CCD and other data, like that for the Office of Civil Rights.

Considering how politically combustible are some of the data states are starting to generate—teacher-student achievement links, anyone?—it will be interesting to see how much support federal and state governments put out to ensure all student data are accurate.

Posted by Sarah D. Sparks at 10:36 AM | Permalink | Leave a comment | Recommend

Anyone who has struggled to get a teenager out of bed on Monday morning—or wake themselves—knows how difficult it can be to make the transition from lazy weekend days to the work-and-school week. Now a new study from the University of Munich suggests that the often-sharp contrast between weekday and weekend sleep habits may also contribute to obesity.

It's pretty well documented that sleep improves academic achievement and athletic performance—and that most students today just aren't getting enough.

Researchers led by Till Roenneberg of the University of Munich analyzed sleep logs of more than 65,000 Europeans ages 10 to 80. They found that weekly average sleep plummets between ages 10 and 20, largely because of increased social and work obligations. More than 80 percent, the researchers found, used an alarm clock to wake during the week, indicating that they had to wake earlier than their natural sleep cycle, and those with a later circadian rhythm, such as teenagers, had even more disrupted sleep patterns.

Most of the study participants tried to pay off their sleep deficit on the weekend, staying out later at night and sleeping and eating later on Saturday and Sunday. This created what Roenneberg and his colleagues called "social jetlag," because the Monday-morning drag was similar in both cause and effect to flying West several timezones on Saturday and flying back again Monday morning:

"The symptoms of jetlag (e.g., problems in sleep, digestion and performance) are manifestations of a misaligned circadian system. In travel-induced jetlag, they are transient until the clock re-entrains. In contrast, social jetlag is chronic throughout a working career."

The researchers found one in three people experienced two hours or more of social jetlag every week, and teenagers suffered the most extreme social jetlag of any age group. Moreover, the greater the disconnect between week and weekend sleep cycles, the more likely a person was to be overweight or obese, and the risk was independent of how much total sleep someone had.

With schools under pressure to help stem the rising child obesity epidemic, this study suggests that rethinking the school day—particularly in middle and and high school—may be as important as getting students moving more at recess.

The study is available in this month's issue of Current Biology, and here's Roenneberg (who has a very soothing voice) talking about the findings.

Posted by Sarah D. Sparks at 10:22 AM | Permalink | Leave a comment | Recommend

The push to promote more "feminine" role models for science, technology, engineering and math fields—think computer engineer Barbie with her pink laptop—may backfire with middle school girls, according to a new study published in Social Psychological and Personality Science.

The study comes in an already-grim week for girls in STEM. Among multiple closing science achievement gaps in the latest results of the National Assessment of Educational Progress, the gap between boys' and girls' average scores widened from four points in 2009 to five points in 2011. In states like Mississippi, Pennsylvania and West Virginia, the gender gap swelled to eight points.

That's still small compared to the 20-to-30-point gap between white students and their black and Hispanic peers, or the equally large divide in performance between wealthy and poor students. However, while both girls and boys performed on average at the "basic" level on the NAEP (149 for girls versus 154 for boys in 2011), there were twice as many boys as girls nationally performing at the highest level in science (2 percent versus 1 percent, respectively).

Any way you slice it, that's disappointing news for the educators and STEM advocates who have been trying to coax more girls to enter science fields. Many recent programs have tried to do so by promoting more women role models in STEM fields; for example, the American Association for the Advancement of Science has papered its recent conferences with posters of women scientists. University of Michigan psychology researchers Diana E. Betz and Denise Sekaquaptewa have found that might be a trickier approach than it seems.

In the first experiment, 144 6th and 7th grade girls read articles about three successful female university students. In some cases these were overtly "girly," wearing pink clothes and make-up and saying they like to read fashion magazines, while in other cases the students wore dark clothes and glasses and simply said they liked to read. The role models also either were specifically described as successful in a STEM field, math, engineering or biochemistry, or were reported as generally successful—for example called a "freshman star."

The researchers found girls who read about the overtly female role models actually reduced the students' reported interest, perceived ability and future expectations in math, and they showed less interest in taking math classes in high school and college than girls who read about role models in more neutral clothing or with non-STEM-specific achievements.

The second study, of 42 6th and 7th grade girls, repeated the prior experiment but first identified how much they liked science and how strongly the girls believed women could be both feminine and successful in STEM careers. The researchers found girls who already disliked science or felt disconnected from it were even more likely to reject the feminine STEM role model than girls who liked science already.

"Submitting STEM role models to Pygmalion-style feminine makeovers," the researchers concluded, "may do more harm than good. A more fine tuned approach is needed to benefit girls with different levels of STEM interest and to protect current STEM self-concepts."

Previous research has suggested that drawing students' attention to sharp gender differences, such as the pink clothes and fashion magazines, may exacerbate students' gender biases. However, in an email, Betz and Sekaquaptewa said that they don't think that's the problem in this case. Rather, they believe the cognitive clash of two gender stereotypes—images of a successful feminine woman and a successful scientist—can convince girls who already feel disconnected from science that they will never measure up to "counter-stereotypic success."

"The bottom line, though, is that this research suggests that we don't need to make role models or STEM fields 'girly' to motivate girls," they told me. "Instead, we should turn to what we already know makes a helpful role model. Girls have to feel like they can relate to or identify with the female scientists they see and learn about. ... Female role models should also be shown as actively involved in science rather than passive observers or tokens: show women really using equipment and conducting research. Teaching girls about what scientists and engineers really do, and especially highlighting their social usefulness and communal aspects, has been found to be motivating for girls."

Posted by Sarah D. Sparks at 12:13 PM | Permalink | Leave a comment | Recommend

The Knowledge Alliance, a Washington-based nonprofit which represents the nation's regional educational laboratories and other research groups, has tapped veteran Hill staffer Michele McLaughlin to be its next president.

McLaughlin, now senior education policy adviser to Senate Education Chairman Tom Harkin, D-Iowa, will start May 29, six months after Jim Kohlmoos left the group to become executive director of the National Association of State Boards of Education.

"She was our unanimous choice," said Max McConkey, the chief policy and communications officer for WestEd and a Knowledge Alliance board member who was on the selection committee. McLaughlin is "well respected, smart, personable, collaborative, with a wonderfully rich blend of experience," he said. "We're excited."

It makes sense for the group to pull in someone deeply connected to Congress, after a grueling year fighting to keep the regional labs system from being axed during Congressional budget fights. Moreover, the newest iteration of the labs includes several new research groups and a new federal emphasis on developing partnerships.

"I am honored to have been selected by the Knowledge Alliance Board to help lead the organization," McLaughlin said. "With the ongoing debate over reauthorization of the Elementary and Secondary Education Act and the pending reauthorization of the Education Sciences Reform Act, I believe there has never been a better time for Knowledge Alliance to lead as the premier policy and advocacy organization for federal education research and to continue to grow in organizational strength."

In addition to her work with the Health, Education, Labor and Pensions Committee since 2010, McLaughlin has held education policy positions at Teach for America, the American Federation of Teachers and the New Jersey education department. She also conducted research on minority student achievement on the National Assessment of Educational Progress, or NAEP, at Princeton University.

Posted by Sarah D. Sparks at 9:20 AM | Permalink | Leave a comment | Recommend

If you missed that April is National Distracted Driving Awareness Month (preoccupied, perhaps?) the National Institute for Child Health and Human Development has released a plethora of research on why teenagers are such risky drivers and ways school districts can help them drive safely.

Teenagers have the highest crash rate of any age group of drivers, and the NICHD's Naturalistic Teenage Driving Study found that they are five times more likely than adults to drive unsafely, and four times more likely to crash or nearly crash.

In a series of studies, NICHD researchers put sensors on the vehicles of 42 Virginia adolescents in the first 18 months after they had received their driving licenses; the trackers recorded "high g-force events" such as sudden braking, hard turns or acceleration. Researchers found that the more such events a teenager had in the last month, the more likely he or she was to get into a crash or near-crash. An adult's presence in the car cut the risk of crashing by 75 percent, but if a driver had other teenagers in the car, particularly boys, the likelihood of crashing nearly doubled. In fact, a car full of teenage boys showed the highest risk of a fatal crash.

The teen driving study found that texting, eating and reaching for objects were the most common things that led to crashes— and while these are distracted behaviors for adults, too, they are much more dangerous for teenagers, researchers found. A teenager dialing a mobile phone was seven times more likely to crash, while for an adult the risk increased only 2.5 times.

A separate study showed why: Adult drivers on a test track who were asked to make a call would glance up as they approached an intersection, spotting a changing yellow traffic light. Teenagers, on the other hand, would complete dialing the number before looking up, thus often running a red light.

In response, the NICHD has provided funding for University of Massachusetts-Amherst researchers who are developing a computer simulation program to train new drivers to spot and respond to road dangers.

Posted by Sarah D. Sparks at 2:07 PM | Permalink | Leave a comment | Recommend

A new atom-based magnetic sensor could help scientists get a clearer look at how children's brains work.

Magnetoencephalography, or MEG, is an imaging method that uses magnets to measure the magnetic fields created by electrical activity when different parts of the brain are activated. It is frequently used to map the areas of the brain associated with different cognitive functions, as well as identify specific signatures related to conditions such as autism and epilepsy.

Yet MEG is notoriously difficult to use with wriggly infants and young children; the scanners use bulky, helmet-shaped arrays of supercooled sensors called SQUIDs—superconducting quantum interference devices—and children often become frightened of the enclosed space or find it difficult to stay still. Only a few research centers in the country have the capacity to use MEG for children's brain imaging, and only one, the Institute for Learning and Brain Sciences at the University of Washington in Seattle, has the capacity to study infants in this way. That has meant that much of what we know from brain research on cognition and sensory perception comes from studies of adults, whose mature brains may respond very differently from those of still-developing children and adolescents.

That's why a new sugar-cube-sized sensor being tested by the National Institute of Standards and Technology in Boulder, Colo., has such promise.

The tiny senor, which successfully measured brain activity in a series of experiments in Berlin last week, uses an infrared laser and fiber optics to measure the light absorbed by 100 billion rubidium atoms contained in a gas. The rubidium absorbs light in proportion to a magnetic field, and can operate at room temperature.

That means such sensors could someday lead to light, flexible MEG helmets that would give children the ability to move around while wearing them, according to Svenja Knappe, a co-developer of the sensor and a physicist with the atomic devices and instruments group at the National Institute of Standards and Technology in Boulder, Colo.

"It will take several years to develop a prototype system that can be used for actual clinical studies, but we are excited about the first steps we made in this direction and the encouraging results that make us believe that our mini-magnetometers can become an alternative to SQUIDs in certain biomedical applications," Knappe said.

The research team still has a ways to go to develop more sensitive measuring devices to use with the sensors. Yet if the mini-sensor matures, it could go a long way toward making brain research more relevant to education.

Photo: The mini-sensor's tiny lens, visible in the sensor, allows a low-energy infrared beam to measure magnetic fields created by human brain activity. Photo courtesy of Svenja Knappe and NIST.

Posted by Sarah D. Sparks at 7:32 AM | Permalink | Leave a comment | Recommend

Sure, we've been hearing about how watching television rots kids' brains for decades now, but apparently secondhand television can be harmful to children who aren't watching it, too.

According to a new media study presented at the International Communication Association annual conference in Phoenix, Ariz., children ages 8 months to 8 years are exposed to nearly four hours each day of television playing in the background.

Matthew Lapierre, a doctoral candidate at the University of Pennsylvania's Annenberg School for Communication, Jessica T. Piotrowski, assistant professor for communication research at the Universiteit van Amsterdam, and Deborah L. Linebarger, an education professor at the University of Iowa, surveyed more than 1,450 English-speaking homes with children from 8 months to 8 years old. The researchers found on average children spent four hours daily with television in the background—not counting the more than 80 minutes that children under 6 watch television shows on average each day.

Television was even more likely to be the soundtrack for young and minority children's lives. Children under 2 years old had background television on average 5.5 hours a day, compared to under 3 hours a day for children 6 and older. Likewise black children were exposed to 5.5 hours of background television each day, compared to 3.5 hours each day for white children.

Prior research suggests background television can have a "chronic disruptive impact on very young children's behavior." Studies have linked background television to less focused play among toddlers, poorer parent-child interaction, and interference with older students' ability to do homework.

"For every minute of television to which children are directly exposed, there are an
additional 3 minutes of indirect exposure, making background exposure a much greater
proportion of time in a young child's day," the study noted.

"Considering the accumulating evidence regarding the impact that background television exposure has on young children, we were rather floored about the sheer scale of children's exposure with just under 4 hours of exposure each day," Lapierre said in a statement on the study. Lapierre and his fellow researchers recommended that parents, teachers and early childcare providers turn off televisions when no one is watching a particular program and that parents prevent children from keeping a television in their rooms.

It's easy to think about this as just one more alarm about how our modern media environment is ruining our kids. Yet the more interesting take-away from this field of research is how critical it is for children to learn actively and socially. Children learn from adults speaking to, with and around them, and from actively engaging with their world.

Anything that limits or distracts from that active interaction can be a problem, but not an insurmountable one. For example, researchers at the University of Washington's Learning in Formal and Informal Environments, or LIFE, Center, is doing some fascinating work on the potential benefits of interactive media. There's also been some interesting work on using video conferencing to read with children.

Television, the internet and other media aren't going to disappear from children's lives, but they do seem to be evolving into more interactive forms. It will be interesting to see whether media will evolve into something more interactive than background noise.

Posted by Sarah D. Sparks at 7:38 AM | Permalink | Leave a comment | Recommend