When the term "research says" or "research supports" comes up in a conversation about education, it is greeted with cynicism.
In part, this is because many confuse the diverse types of research. The purpose of this brief essay is to articulate a way of looking at research studies and bodies of research that clearly delineates different types of research and the very different implications for practice. This trichotomy of research will serve as a framework for determining what research relates directly to validated practice. (Ellis & Fouts, 1993) Although there is some overlap between categories, most empirical studies in the field of learning disabilities seem to fit one of the three. The following is a potential framework to use to "sort out" types of research:
The first research category is descriptive.
Descriptive research can utilize either qualitative or quantitative methodologies. These studies can be very useful for theory building, for helping shape interventions, and for helping understand the target or focus of an intervention. For example, descriptive research comparing the actual performance of proficient readers with the performance of children with learning disabilities served as a starting point for highly effective interventions in the areas of reading comprehension and expressive writing. These studies can help us understand the nature of problems faced by educators, e.g., focusing efforts on phonemic deficits in students with reading disabilities, or designing and evaluating the effectiveness of interventions geared towards helping students with learning disabilities address their organizational/metacognitive problems. Descriptive studies can also help us understand common implementation problems and other pressing problems in current practice. However, despite the rich insights they often provide, they can not serve as evidence of effectiveness.
The second type of research is well-controlled experimental and quasi-experimental research studies.
These studies are the building blocks of scientific knowledge about teaching and learning. These studies allow us to see, for example, what students with learning disabilities can learn when taught by excellent teachers using state of the art methodologies. They can - and should - guide professional development and curriculum development. Although decades ago these studies were so tightly controlled that they had a laboratory-like feeling to them, this is rarely the case with contemporary research. Usually these studies are now conducted in schools and reflect many of the realities of current practice.
The third type of research involves large-scale field studies.
Often, but not always, these involve multiple sites and are longitudinal in nature. Examples of these are Project Follow Through, the longitudinal research on classwide peer tutoring by Greenwood and associates, the evaluations of bilingual education by Ramirez and Danoff, and the research on early reading by Foorman and associates. These studies provide realistic "field tests" of what works. Typically, the internal validity of these studies is weaker than small-scale research typical of the second category. However, they are strong in terms of external validity for several reasons. Because so many teachers and children are involved, level and quality of implementation typically runs the full gamut. Thus, they often provide "tests" of what can work in the real world of schools. Also, the large sample sizes increase confidence in findings.
It is important to discriminate this type of research from routine public school evaluation studies, which typically lack adequate control.
All types of research are critical for the development of valid knowledge in a field. It makes sense for the educators to focus on the latter two types, since these are the only two types of research that relate directly to validated practice.
Ellis, A., & Fouts, J. (1993). Research on educational innovations. Princeton, NJ: Eye on Education.