Why Do We Look for Pebbles?

As program chair for the special interest group Research on the Education of Deaf Persons of the American Educational Research Association (AERA), I organized a special session for the 2005 annual meeting in Montreal entitled "Pebbles in the Mainstream." In that session, Michael Karchmer of Gallaudet discussed the current demographic trends in Deaf education, Marc Marschark of the National Technical Institute for the Deaf (NTID) spoke about changes in publishing practices, and Ronald Kelly of NTID tackled the future of research methods in our field. While it was apparent from the session that the business of educational research on programs for deaf or hard of hearing children was in considerable flux, no one was prepared to take on the issue with much confidence. Nonetheless, it seemed that it was an issue that needed some consideration.

One truly bright spot at the AERA meeting was Stephanie Cawthon's study of the response of states to meeting the testing mandates of the No Child Left Behind Act when it applied to deaf or hard of hearing children. She was clearly addressing a very critical and timely issue using some innovative methods. Essentially her approach amounted to a marriage of the qualitative method of "snowball" sampling to the technology of e communications. Cawthon's paper was encouraging for a couple of reasons: First, she represented the "future" of the field, that is, a recent PhD recipient; and second, she was willing to look into new methods of collecting data.

This realization led to a consideration of other current or recently completed PhDs who were trying to deal with new methods or who were trying to make old methods work in a new professional environment.

Kelly Crain, who recently joined the faculty of Florida State University, encountered numerous methodological challenges in his dissertation work when he tried to do a classic causal comparative or quasi-experimental study. His dissertation on phonemic awareness among Cued Speech users reflected the problems posed by Karchmer's description of a smaller, more scattered, and more diverse population of deaf or hearing children than existed 20 years ago. Crain's experiences led him to question even the possibility of doing experimental research in the field of education of deaf or hard of hearing children given the near impossibility of generating a truly random sample.

At the same time, Connie Morris had just left Gallaudet University to take a position at the Georgia School for the Deaf in Cave Springs. Morris had been part of the same cohort of PhD candidates at Gallaudet as Crain. One of her last jobs at Gallaudet had been working with me on a secondary analysis of large databases. This was a fun and frustrating experience because every time we thought we had a rationale for identifying the deaf or hard of hearing children, the numbers seemed to vanish; or if we loosened our criteria, the numbers jumped, making us question the validity of our sample definition.

This issue of the American Annals of the Deaf is not intended to provide answers about the future of research on the education of deaf or hard of hearing children; nor does it offer solutions to the problems it raises. The goal of these articles is to continue a dialogue begun in Montreal in April 2005: Does the future of research in our field lie in methods that while internally valid offer no external validity, or are there methods of research that will recognize the challenges of our field while yielding results that are both internally and externally valid?