A set of in-depth partnerships between researchers and the nation鈥檚 large school systems are bearing fruit, both in new studies and greater insights into how districts and researchers can work together.
The released two reports from the second of three rounds of studies this month in Washington at the annual legislative conference of its sponsor, the Council of the Great City Schools. Researchers associated with Washington-based Strategic Education Research Partnership or SERP are developing tools and interventions in Boston and San Francisco to help middle and high school teachers, particularly those in science, social studies, and other content areas, incorporate academic vocabulary into their teaching.
One report describes efforts in the 56,300-student Boston district to develop , a series of 15-minute daily academic-vocabulary lessons and activities that could be incorporated into different courses. With the CGCS grant, Catherine E. Snow, the head field researcher for SERP and an education professor at Harvard University, and Harvard postdoctoral fellow Joshua F. Lawrence dug into data on effectiveness for different student groups.
The researchers found that students in the Word Generation schools outperformed students in comparison schools on end-of-year tests of academic vocabulary. Moreover, English-language learners in participating schools outperformed English-proficient peers at the control schools.
Online Search Tool
The second report, on the 56,000-student San Francisco district, discusses the efforts of science teacher Lisa Ernst and six other teachers who have been working with Kenji Hakuta, an education professor at Stanford University, and others from SERP to develop WordSift, an online search tool that allows students and teachers to connect academic words or paragraphs to related words, videos, and sample text explaining the concepts.
鈥淐ontent teachers tend not to think of themselves as responsible for the literacy development of their students, so if you ask them to think of vocabulary, they will suggest actual [subject-matter] content words like 鈥榦smosis,鈥� 鈥� Mr. Hakuta said. 鈥淏ut there鈥檚 another class of words and aspects of language that cuts across content, academic language such as 鈥榓nalyze鈥� or 鈥榚xplain,鈥� which doesn鈥檛 usually show up in everyday conversation but is a very regular part of academic vocabulary.鈥�
Research has shown that academic language is critical for students鈥� ability to understand and communicate academic content, particularly in upper grades.
Mr. Hakuta said the tool now gets between 500 and 2,300 users a day, and he has started evaluation trials.
鈥淚 think [regarding] the development [of WordSift] itself, many of the teachers saw it as professional development and said it was the best PD they鈥檇 ever had, because they got to co-develop it,鈥� Mr. Hakuta said. 鈥淚t wasn鈥檛 intended as professional development for the teachers, but they saw it as that.鈥�
Yet that has been a goal of the partnership project, according to Amanda Rose Horwitz, the research manager for the Council of the Great City Schools.
鈥淚 feel urban districts often feel poked and prodded; they feel like the subjects and not partners in the studies,鈥� Ms. Horwitz said. 鈥淎t the same time, researchers are just throwing themselves up against these walls. Large urban districts can be amazing bureaucracies sometimes.鈥�
Looking Long-Term
The Washington-based council, which represents the 66 largest school districts in the country and 15 percent of K-12 public school students, launched the research fellowship in 2006, with support from the U.S. Department of Education鈥檚 Institute of Education Sciences. The fellowships provided up to $150,000 each for 18 months for research staff and supplies in nine school systems. The CGCS grants are intended to support the development and continuation of long-term partnerships like those already begun by SERP, an initiative originally launched by the National Academies, and the Value-Added Research Center, or VARC, at the University of Wisconsin-Madison.
In each city, the district sets research priorities, and scientists and district administrators develop a portfolio of research projects together. For example, Geoffrey D. Borman, a professor of educational leadership and policy analysis at the University of Wisconsin-Madison, paired with the St. Paul, Minn., schools for the council鈥檚 next round of studies. In exchange for evaluating professional learning communities in the schools鈥攁 top district priority鈥擬r. Borman will have district support to test a series of writing exercises intended to combat stereotype threat in minority students.
As the researchers and districts work on their studies in each partnership, the council has been culling best practices in how to develop such research partnerships, which have become a top priority for the ies.
鈥淲e find these really viable research partnerships are really popping up in a lot of different places,鈥� Ms. Horwitz said.
Mr. Hakuta of the San Francisco project said working in an ongoing partnership gives him a different research perspective.
鈥淭ypically, if I go to a district, I might have a problem in mind that I want to address, and I might shop it around,鈥� he said. 鈥淚 might be trying to look for answers to problems they don鈥檛 have.鈥�
Burrowing In
In Milwaukee, the site of one of the , researchers from the University of Wisconsin鈥檚 VARC center have 鈥渆mbedded鈥� two researchers in the district鈥檚 research office. While both are university researchers, one works two days a week and one five days a week on campus.
Deb Lindsey, the Milwaukee school district鈥檚 director of research and evaluation, said the partnership has saved the 82,400-student district money by providing experienced researchers with access to the university鈥檚 resources. Yet, more importantly, she said, it has given the school system researchers 鈥渄riven by a genuine interest in helping the district learn and improve.鈥�
鈥淭hese are not people coming for a master鈥檚 thesis; it鈥檚 not a publish-or-perish situation,鈥� Ms. Lindsey said. 鈥淭he ideas are co-developed; they don鈥檛 say, 鈥業 see you have a problem there, and this is how I think we should study it, or this is how you should fix it.鈥� 鈥�
In the process, the team has developed several program evaluations and a districtwide early-warning data system to spot students at risk of dropping out of school, according to Bradley Carl, a VARC researcher embedded in Milwaukee public schools. The district continued the partnership after the Great City Schools grant ran out, and the team is now piloting an expanded warning system to identify students who may graduate but encounter problems attending and completing college, he said.
Robert H. Meyer, VARC鈥檚 director, said: 鈥淣ormally, the people who would be good at the technical risk system wouldn鈥檛 be good at the reporting portion, or might not be good at developing interventions, or evaluation. If you weren鈥檛 in embedded research, you might try to do one of those portions and then hope that someone else picks up the other parts.鈥�
The success of the project has led Mr. Meyer, a research professor at the University of Wisconsin-Madison, to expand the embedded-researcher program in partnerships in Atlanta, Chicago, Hillsborough County, Fla., Minneapolis, New York City, and Tulsa, Okla.
