Researchers have found that, in the right circumstances, it is possible for students from special populations to complete an online class and demonstrate as much academic learning as those who completed a face-to-face course.
They caution, however, that using that conclusion to justify any particular online learning option for special education, English-language-learner, at-risk, and gifted students in need of an intervention is perilous, at best.
鈥淲e don鈥檛 really know at this point which [methods] tend to be the most effective鈥� for specific populations, said Cathy Cavanaugh, a professor of educational technology at the University of Florida, in Gainesville, who was co-author of a study about improving outcomes for virtual school students with disabilities that was published last year. 鈥淲e need to get to a much more fine-grained level of understanding of what鈥檚 going on,鈥� she said.
Yet the small cadre of researchers who focus on online K-12 education believe making their case to the public for more nuanced research, as well as gaining support and funding for it, can be difficult, says Michael K. Barbour, an assistant professor of technology at Wayne State University, in Detroit.
鈥淭he problem is, most academic journals, most policymakers, and most people in the press want to find usable nuggets that can be used in other situations,鈥� said Mr. Barbour, whose research includes a 2008 dissertation exploring rural online education in Newfoundland and Labrador, in Canada.
鈥淲hat we need is not large, generalizable studies,鈥� he said. 鈥淲hat we need is for research and evaluators to work with programs in a design-research model.鈥�
Mr. Barbour insists there are significant imperfections inherent in most broad studies of online learning for special populations and, indeed, in studies of online learning in general.
For example, he said, some past studies have sometimes compared the achievement data of virtual students who chose virtual learning on their own against a general sampling of brick-and-mortar students. Other studies鈥攕uch as research that has found students with autism perform better on standardized tests after receiving online instruction鈥攈ave used a narrow measure of educational quality based largely on standardized-testing results, he said.
Mr. Barbour argues that shifting from wide-ranging, short-term studies to narrower, more in-depth explorations designed to solve a problem at a specific online program would be a more effective method of research. And with wide disparities between how special populations of students are educated in online programs across the nation, he says, narrower research may provide more meaningful feedback than broader studies that attempt to make comparisons of dissimilar data.
鈥淚f I can solve a problem with Michigan Virtual School,鈥� he said by way of example, 鈥渁t the end of the day, I鈥檓 going to come up with a conceptual framework, a theoretical framework, to explain why what we did eventually worked.鈥�
Uniform Data Scarce
A lack of data uniformity is one obstacle preventing apt comparisons, says Richard E. Ferdig, a professor of research and information technology at Kent State University, in northeastern Ohio. He is the founder of the Virtual School Clearinghouse, a collaborative research project designed to collect data on the field.
The field of research regarding online education for special populations is still developing. Here is a summary of some early research on the subject.
U.S. DEPARTMENT OF EDUCATION
This meta-analysis of recent experimental and quasi-experimental studies contrasting online and blended learning against face-to-face equivalents found blended learning to be more effective than either purely online or purely face-to-face. But it did not find online course delivery as a medium superior to face-to-face delivery. The review also notes that many studies suffered from weaknesses such as small sample size, failure to report retention rates in the scenarios being contrasted, and potential bias coming from experimenters doubling as instructors.
INTERNATIONAL ASSOCIATION FOR K-12 ONLINE LEARNING
This issue brief gives results from a survey of virtual schools that was conducted to determine how many of the estimated 175,000 full-time online learners fell into the 鈥渁t risk鈥� category; the researchers found that nearly half the responding virtual schools had full-time student populations in which at least 50 percent of the students were deemed to be at risk academically in some fashion. The brief examines how schools attempted to meet the nonacademic needs of at-risk students at a program level, as well as how the schools designed and delivered instruction for those students.
ASSOCIATION FOR THE ADVANCEMENT OF COMPUTING IN EDUCATION
This report explores the experiences of rural students in the Canadian province of Newfoundland and Labrador who enrolled in courses offered by the province鈥檚 Centre for Digital Learning and Innovation. It found that students鈥� experiences could be made better by improving teachers鈥� asynchronous鈥攔ather than real-time鈥攐nline instructional methods. Unlike some virtual schools, courses at the CDLI contained both synchronous and asynchronous components. The report also recommended developing ways to get students more socially involved through virtual means, and making students more aware of the CDLI鈥檚 support resources. The report also stresses that most of its findings would not be applicable on a large scale to other virtual schools because of their varying structures.
QUARTERLY REVIEW OF DISTANCE EDUCATION
This report identifies five principles that it deems crucial to serving students at risk of dropping out and explores ways different virtual schools are incorporating those principles into their school cultures. The principles, which the report calls the 鈥渇ive C鈥檚,鈥� are: 鈥渃onnect, care, climate, curriculum, and control.鈥� For schools to better serve such students, the report includes in its recommendations creating an online/blended teaching endorsement to add to a teaching license, professional development specifically geared to online instruction of at-risk students, and the development of online-teaching standards for special education instructors.
鈥擟ompiled by Staff Writer Ian Quillen
For example, Mr. Ferdig says, getting hold of data from virtual schools is a problem, because most online schools serve students as a supplement to those students鈥� coursework at regular schools, and report academic data to those schools.
And in many cases, he says, virtual schools鈥� identification of students who make up special populations is imperfect at best. Even though research suggests virtual schools are serving a greater proportion of students with special needs (except for English-language learners) than brick-and-mortar schools are, those students and their parents may not identify themselves as such when they enroll. In fact, many often embrace a virtual option to avoid identification as one of those student population groups.
鈥淭here鈥檚 this paradox,鈥� said Mr. Ferdig, who has researched online education as a measure for reaching students at risk of dropping out of school, most recently with work published in 2010 that studied such interventions at the Michigan Virtual School, which serves courses to about 16,000 students. 鈥淵ou want to give kids the support they need,鈥� he said. 鈥淏ut at the same time, you also want to value the anonymity online learning provides.鈥�
Structure and Organization
Leanna Archambault, an assistant professor of education at Arizona State University, in Tempe, says because virtual education is still new and developing, research also needs to expand its focus to solve questions of structure and organization.
Those questions include how virtual schools can construct intervention programs that mirror those that reach out to English-language learners, at-risk students, gifted students, and students with disabilities in brick-and-mortar schools. And it also includes educating teachers who will be ready to deal with those challenges in a virtual environment.
鈥淭he questions are not only concerns on an instructional level, but on a programmatic level, and at the school level [about] what needs to happen,鈥� said Ms. Archambault, who is hoping to combine her past exploration of at-risk online learners with her expertise in teacher education.
鈥淗ow do I prepare a teacher? What are skills a teacher needs to work with an at-risk student in an online environment? Those are huge variables,鈥� she said.
Ms. Cavanaugh, of the University of Florida, said it appears that policymakers and others are warming to the idea that online programs should be an option for special populations of students.
For example, a of more than 400 principals found that most expressed interest in using online education for credit-recovery courses in their schools, even while holding reservations about the quality of those courses. , released last December, which advocated changes to state policy to give all students more options to online learning, was produced by a collaboration headed by former Govs. Bob Wise of West Virginia and Jeb Bush of Florida, not education researchers.
Ms. Cavanaugh hopes that shift in thinking will allow for more nuanced and thorough study, and in turn attract more researchers to what is still a lightly populated field.
鈥淭he virtual-school research community is small and certainly needs collaboration,鈥� she said. 鈥淢y hope is people who have been studying mainstream education will begin to recognize the scope of virtual education and bring their expertise to this field.鈥�
