A final evaluation of the federal Enhanced Reading Opportunities program suggests that extra, explicit reading classes can boost reading skills for struggling adolescents, but the short-lived improvements aren't enough catch up students years behind the curve.

Still, the last of three reports by the New York-based research firm MDRC Inc., released today, provides clear, positive results from a large-scale randomized, controlled trial on reading intervention &mdash a rarity in the increasingly urgent debate over adolescent literacy.

The study, conducted on behalf of the U.S. Department of Education's Institute of Education Sciences, suggests an extra reading class instead of a regular elective can boost reading comprehension, GPA, course credits and even state reading and math scores for students who enter freshman year reading several grade levels behind.

MDRC researchers tracked 6,000 9th-grade students in 34 high schools who read at least two years below grade level. The students were randomly assigned to attend either one of the two supplemental reading programs &mdash Reading Apprenticeship Academic Literacy, developed by San Francisco-based WestEd researchers, and Xtreme Reading, a brainchild of the University of Kansas Center for Research on Learning in Lawrence &mdash or a scheduled elective course. At the end of a full year of classes that ran 225 minutes each week, all three groups of students took a standardized reading comprehension test. In addition, both during the treatment year and the following year, the researchers looked at students' end-of-year course credits, performance in state assessments of reading, math, science and social studies, as well as behaviors such as school attendance and reading both in school and on their own time.

Researchers found students in the demonstration programs moved from the 23rd to the 25th percentile in reading during the year, representing about two months of growth more than their peers in the control group. In addition, the demonstration-group students had a 13 percent higher average GPA, and performed better than the control group on state assessments of both reading and math during the program year. Demonstration participants also accumulated more credits toward graduation than their peers.

"It provides some glimmer of sense that there was some transferability of the literacy skills," said Marsha Silverberg, IES program officer on the study. She noted that both programs improved students' reading comprehension, but only the Reading Apprenticeship did so significantly.

So why am I not doing a happy-dance right now? Because the gains, interesting as they are, weren't nearly enough to make up the difference for kids who started out on average four or five years behind grade level in reading. At the end of the year of supplemental help, nearly four out of five students still read two or more years below grade level and neither they nor the control group were on track to graduate. By the next year, every one of those program-year gains had disappeared. Moreover, even during the program year, students didn't improve their reading behaviors along with their reading performance: they didn't report reading more often, attending school more often, or even using the strategies taught in either of the two programs.

"I think it's really important to think of how low-level a group these students were," Ms. Silverberg said. "When we put out the call, we said [participants should read] at least two years behind, but I would think the students we ended up with had even lower reading skills than they expected. Remember, literacy doesn't get taught in high school, it doesn't even get taught in middle school. Explicit literacy basically stops in 5th grade. Is one year of a program enough to get these students on track to be adept readers when all their lives they weren't adept readers?"

Moreover, course material changes dramatically from class to class in complexity and content during the high school years, said Rich M. Long, government relations director for the International Reading Association. "We have evidence now that [students] do learn and they do generalize from the literacy class to the content class, but when the material gets harder, the improvement isn't sustained," Mr. Long said. "The effects of the environment &mdash poverty and the like &mdash don't dissipate with an infusion of specific skills. You still need gas to run the car."

Recent studies like the MDRC reports contribute to a solidifying research base on what works in adolescent literacy, Mr. Long said, but the rapid increase in rigor for "college and career readiness" standards makes the need for such evidence at the high school level more urgent.

Literacy in secondary schools, Mr. Long said, is "much harder than elementary schools [because] high schools tend to be a whole lot bigger, the issues and complexities of what is being taught tend to be a whole lot more daunting. Just having an emphasis in elementary schools doesn't mitigate against the effect of poverty, especially in a time when we're making things more rigorous across the board."

Fourteen out of the original 34 study schools have chosen to keep the supplemental reading programs this year, with some extending the time of the program to allow students to take the supplemental class for more than a single year. Though there are no plans to continue tracking the students' progress in those schools, Ms. Silverberg said IES does hope to use the lessons learned from this study to help theeight grantees in this round of President Obama's Striving Readers adolescent literacy program.

Posted by Sarah D. Sparks at 9:39 AM | Permalink | Leave a comment | Recommend

If you're feeling refreshed after the long holiday weekend, here's a meaty new report to tackle: The Paris-based Organization for Economic Cooperation and Development has just released its Education at a Glance 2010, a bit of a misnomer for the 470-page tome.

Education at a Glance provides a buffet of more than 100 indicators on everything from international teacher salaries (still below comparable fields in most countries) to school choice (on the rise, with more than half reducing barriers to choice among public schools and 10 countries providing new choice money.) It covers information from the 31 mostly European OECD member countries, as well as eight non-members including Brazil, China, India and Indonesia.

This year the OECD seems to echo President Obama's push to improve college graduation rates, with a highlighted section connecting overall average education levels to global unemployment rates during the economic crash. While students who did not complete secondary school have seen unemployment rates topping 9 percent in the past year, only 4 percent or fewer of those with a college degree (which the OECD calls a tertiary education) have faced the same problem.

"With the worldwide recession continuing to weigh on employment levels, education is an essential investment for responding to the changes in technology and demographics that are reshaping labor markets," said OECD Secretary-General Angel Gurría in a statement on the report's release.

I'll be digging into the report throughout the week; in the meantime, let me know your thoughts on any of the indicators: What surprised you?

Posted by Sarah D. Sparks at 7:29 AM | Permalink | Leave a comment | Recommend

The Department of Education just handed out $330 million in grants to two state coalitions to design the "next-generation" tests of students' readiness for college and careers. It's gotten me thinking: What would a perfect test look like? Would "teaching to the test" be kosher if schools had fantastic assessments of higher-order thinking, problem-solving and the like?

In the latest issue of the Journal of Experimental Psychology: Learning, Memory, and Cognition, a study out of Washington University at St. Louis, Mo., suggests students can learn better by teach with the test, rather than to it.

Psychologist Andrew C. Butler, author of the study, found that repeated and varied testing helped students transfer their learning to new tasks better than simply studying the information. In four separate experiments, undergraduate students studied a series of six passages on different topics. For two topics, they repeatedly restudied the material; for the other four topics, the students repeatedly took either the same or varied tests on the material. A week later, the students took final tests on the topics.

Butler found students who had been retested either with the same questions or variations performed at least twice as well on factual questions and nearly 50 percent better on conceptual questions in the final test than did students who studied the material repeatedly. Moreover, in a follow-up experiment, Butler found that students who were tested repeatedly could make better inferences about new questions based on their previous knowledge, such as relating differences in the wings of birds and bats to the maneuverability of new aircraft. Butler theorized that test-taking allowed students both to apply new knowledge and to get feedback to correct misconceptions.

The results emphasize the importance of high-quality tests as a tool for teaching rather than a distraction from it. The National Testing Survey in progress at Harvard University Graduate School of Education is attempting to crystallize an "ideal test," of this sort, according to Zachary Stein, Harvard doctoral candidate and deputy director of the Developmental Testing Service, which sponsors the survey. It's in the early stages, but so far suggests that while people of different ethnic backgrounds and education levels view today's tests differently, there is general agreement about high-quality tests. In short, everyone wants a test that can measure student progress while also helping educators teach.

If you want to put your own two cents in, the survey is still ongoing here.

Posted by Sarah D. Sparks at 8:38 AM | Permalink | Leave a comment | Recommend

The National Center on Education Statistics has tabled its primary study of adult literacy in favor of a smaller, more targeted international assessment intended to gauge Americans' skills in a technologically advanced workplace. The move is part of the center's effort to rebalance its "portfolio" in the wake of budgetary changes and new administration priorities.

"Five years ago we were using different skills in the office than we are now," explained Val Plisko, associate commissioner for NCES' Early Childhood, International, and Crosscutting Studies Division. "It's intended to provide a broad brush indicator of how the U.S. relates to other countries."

The new Program for International Assessment of Adult Competencies, or PIAAC, will roll out with a field trial of 1500 adults in 20 states, before the full assessment of about 5,000 Americans ages 16 to 65 in 2011. The results will be compared with equal groups in 27 other countries. While NCES' current tracker, the Adult Literacy and Lifeskills Study, or ALL, also allows international comparisons, PIAAC will be the first conducted via computer, incorporating what International Activities Program Director Dan McGrath called "problem-solving in a technology-rich environment."

For example, a participant will have to create and use a data spreadsheet to answer questions. The basic literacy questions will also be at "the higher end of literacy," Mr. McGrath said, noting that the computer-based assessment will allow for more critical thinking test items.

Andrew Tyskiewicz, president of the Commission on Adult Basic Education, praised the new questions. "Not that technology is the only set of skills we have to assess, but the reality is there are certain sets of professional skills that are performance based," Mr. Tyskiewicz said. "In the adult education field, the more you contextualize it, the better. We're all for that."
That said, the assessment's once-a-decade schedule and relatively small sample size are causing some consternation among adult education experts, particularly since NCES is not planning a bridge study between the two assessments now.

"Over the years, we rely on a lot of that data for planning and awareness activities to make that known to the general public," Mr. Tyskiewicz said.
It remains to be seen when and how the two assessments will be integrated. Mr. McGrath said NCES may attempt to administer at least the questionnaire portion of the assessment every five years, to shorten the gap between new information, but, he added, "That's completely up in the air."

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

The National Board for Education Sciences, the advisory group which supervises the Education Department's Institute of Education Sciences, has announced the agenda for its next meeting, Sept. 29. Now it has to hope members will be confirmed in time to show up.

At the September meeting, IES Director John Q. Easton plans to discuss revised research priorities for the institute, which the board must vote to approve. The board also plans to discuss the What Works Clearinghouse and a recently released evaluation of charter schools.

The board consists by law of 15 voting members, but had been working with only six until the Senate confirmed four new members in June. Yet the Senate Health, Education, Labor and Pensions Committee has yet to vote on President Obama's four remaining NBES nominees: Anthony Bryk, president of the Stanford-based Carnegie Foundation for the Advancement of Teaching for a term ending in November 2015; and for terms ending November 2012: Robert A. Underwood, former U.S. House delegate for Guam and president of the University of Guam; Kris D. Gutierrez, education professor at the University of Colorado at Boulder; and Atlanta Public Schools Superintendent Beverly L. Hall. Moreover, the committee also hasn't approved Sean P. "Jack" Buckley, of New York, tapped to be the new Commissioner of Education Statistics.

The nominees must get through committee and a full Senate vote during the Senate's 10 business days before the meeting.

Stanford University economist Eric Hanushek's tenure as board chairman is due to expire at the end of November, so he said he may try to squeeze in an additional meeting of the board in early November to smooth the transition for new members.

Posted by Sarah D. Sparks at 9:06 AM | Permalink | Leave a comment | Recommend

In interesting research news, check out my fellow blogger Steve Sawchuk's Teacher Beat analysis of the Center on Reinventing Public Education's recent study of "mutual consent" teacher staffing systems.

These systems often get tapped when districts move away from seniority-based hiring and transfers. Yet the authors Betheny Gross, Michael DeArmond and Dan Goldhaber found the shift from seniority to mutual-consent alone did not reduce teacher turnover or improve the concentration of inexperienced teachers at poor and high-minority schools. I wonder how education reformers are reacting to the Matrix-like fall of yet another silver bullet?

Posted by Sarah D. Sparks at 11:46 AM | Permalink | Leave a comment | Recommend

What does it look like when research evolves out of classroom experience? There's a bit of a philosophical disagreement in the education research community: those who say action research lacks rigor and those who argue lab research lacks relevance. Yet researchers who draw on classroom experience can bring interesting perspectives and enthusiasm to their work.

Here's a good example. Tamar Kushnir, director of Cornell University's Early Childhood Cognition Laboratory came into education research after teaching preschool. Kushnir said a course at the Barnard College Toddler Center ignited her interest in research.

"It was eye-opening for me," Ms. Kushmir recalled. "It was amazing for me because if you don't know how to observe children, it's hard to see them as having any knowledge at all. But when you observe them and get down on the floor with them, I started to realize that ... in 10 minutes of a child's life, there's so much activity going on. They're learning constantly."

Years later, Ms. Kushmir's latest study, published in the July issue of Psychological Science, finds that preschoolers use statistical cues to understand others' behavior and preferences. Prior research has found children as young as 11 months expect a sample of toys drawn from a box to be 'random,' that is, to generally match the proportions of objects in the box, and are surprised if the contents differ. Ms. Kushmir and her fellow researchers Fei Xu and Henry M. Wellman studied 72 preschoolers and found that they noticed when a person chooses a non-random sample and used this to determine the person's preferences. The children's conviction about another's favorite toy was in direct proportion to the likelihood that the person could have chosen five of the same kind of toy at random from a box of available toys.

The study has no direct lessons for educators in teaching young children empathy &mdash or statistics, for that matter &mdash but Ms. Kushmir intentionally modeled the experiment after the daily activities of a preschool classroom, and she said teachers should pay attention to the way even young students experience their classroom.

"I would say what the experiments really teach us is how much students learn without us teaching anything," she said. "Even before the age of 5 they know a lot of causal relationships. They can't tell you what gravity is, but they can certainly tell you that if you knock something over with liquid in it, it's going to end up in a mess on the floor. Even when they're not able to verbalize these concepts, they're engaging the information."

Posted by Sarah D. Sparks at 11:12 AM | Permalink | Leave a comment | Recommend

Five years ago today, the levees broke in New Orleans and we all learned just how unprepared our cities and schools and families were to face a large-scale disaster. For the anniversary, two new studies suggest we're not much more prepared today.

Save the Children's second annual report card on state practices found only 12 states have basic safeguards in place to protect children during similar disasters. The group queried states on whether they require all schools and licensed child-care providers to have four elements in place the organization considers essential to emergency planning for children: a written evacuation plan to move children at the school or center to a safe location; a written plan to notify parents or guardians of the emergency and reunite them with an evacuated child; a written emergency plan that includes procedures for students with special needs; and a school emergency plan that accounts for different types of disasters the school might encounter, including natural disasters or terrorist acts. The 12 states with all those elements in place represent an increase from 2009, when only seven met that criteria. States most often left out requirements to plan for protecting special-needs students during emergencies; only 18 states included that element.

Meanwhile the National Commission on Children and Disasters released the executive summary for its forthcoming report to the President and Congress. The slew of recommendations hints at a lack of coordination among federal agencies around how to anticipate children's needs during emergencies. In particular, the commission calls for better "mental health first aid" to help children cope with trauma and federal support to help closed schools reopen more quickly and receiving schools deal with influx of relocated students. I'll be eager to dig into the details of the commission's findings when the full report comes out in October.

In the meantime, there's some good news: The Department of Education just announced $28 million in Readiness and Emergency Management for Schools grants to 98 districts in 28 states.

Posted by Sarah D. Sparks at 9:50 AM | Permalink | Leave a comment | Recommend

As education watchers reflect on the fifth anniversary of the hurricanes that devastated Gulf Coast schools, they may find a few lessons in how to make all research more relevant to schools.

Back in February, at an Atlanta conference of the American Association of Colleges of Education, the director of ED's Institute for Education Sciences rolled out his "Five Big Ideas" to improve education research. "If researchers want their work to be relevant," John Q. Easton said then, "we need to spend time in schools talking with administrators and teachers about the challenges they face; we need to reach out to policymakers; we need to collaborate with researchers outside their expertise."

Education researchers have complained for years that accountability has made schools increasingly unwilling to open their doors. Yet that normal reluctance was nothing compared to what researchers faced to study schools in the aftermath of Katrina, when administrators were too busy trying to locate students, find homes for returning staff and open their doors to spare time for researchers.

The researchers that got a foot in the door after Katrina went a step beyond simply being relevant.

"We didn't go into schools to gather research data; we went into schools to see how we could be helpful with the recovery," said Joy D. Osofsky, professor of pediatrics and psychiatry at Louisiana State University Health Sciences Center in New Orleans.

Osofsky and her team adapted a mental health and trauma screening tool for 19 schools in Orleans, St. Bernard and Plaquemines parishes. "We would be unobtrusively around, and what they discovered &mdash and we had anticipated &mdash was there were a lot more problems than they had expected. What happened very quickly in all of the schools we worked with was they realized they could get a lot of help and services from us."

Other recovery researchers echoed Osofsky's advice. For example, Deborah M. Alvarez, assistant professor of English at the University of Delaware, led groups of her students to volunteer in the schools and held training seminars for teachers on how to help students improve their literacy skills and mental health by writing about the disaster. Those teachers later helped Alvarez get clearance to work in the Recovery School District, which she said was previously "almost impenetrable" to her as a researcher.

Today, Osofsky's team has five years of longitudinal mental health data on 23,000 students, and it was able to adapt its trauma screening tool and training for educators in Chile after the earthquake earlier this year.

The relationship has also given Osofsky's team an up-close view of how St. Bernard and Plaquemines parish students cope with a new disaster, in the form of the oil spill that is devastating the communities' fishing and tourism industries.

"It's very important to go and hang out, so people just start to trust you, rather than see you as a researcher just going in and collecting data," Osofsky said. "We're sort of embedded in the schools now; we're collecting data, but we're also really a part of these schools at this point."

Posted by Sarah D. Sparks at 3:58 PM | Permalink | Leave a comment | Recommend

While most studies of student mobility have focused on problems in poor urban centers, a recent analysis of five states by Mid-continent Research for Education and Learning, or McREL, suggests rural mobility rates can rival that of urban districts.

Frequent school changes can have a huge effect on a student's academic career, from low grades and test scores to behavior problems and difficulty making friends. Moreover, high student mobility can break down the learning process in a school as a whole, as teachers devote extra energy to helping incoming students catch up, and administrators try to provide teachers, space and materials for shifting numbers of students. In a small rural district, it takes a lot fewer kids moving around to play havoc with your staff assignment, special education supports and even course offerings.

Andrea Beesley, McREL senior director and lead author on the report, studied state-reported student mobility data for Colorado, Missouri, Nebraska, North Dakota and Wyoming. The research team found that in Wyoming and North Dakota, rural districts had higher student mobility than did cities, though the smaller school populations in these districts may skew the sample.

Moreover, rural districts, particularly those considered remote by National Center on Education Statistics definitions, often showed "extremely high" mobility, defined as rates two standard deviations or more above the state average.

"Hotspots" of high student mobility often coincided with districts with high poverty or Native American reservations. Yet much more than that we can't tell, because each state collects and reports its data on student mobility differently: different grade levels included, different ways to include students who leave and return several times, and practically no information on where a mobile student comes from or goes to when he leaves. None of the states differentiated the highly mobile migrant student population from regular mobility counts.

"No two of them do it the same way," Beesley said. "Unfortunately it leaves you in the situation where in many cases you don't understand a lot about student mobility.

The team hopes to expand its research as state longitudinal data systems evolve and high schools begin to implement the Title I requirement to track student graduation rates longitudinally. It will be interesting to watch the picture on these rural schools come into focus.

Posted by Sarah D. Sparks at 9:59 AM | Permalink | Leave a comment | Recommend