Kirk and Miller define what is -- and what is not -- qualitative research. They suggest that the use of numbers in the process of recording and analyzing observations is less important than that the research should involve sustained interaction with the people being studied, in their own language and on their own turf. Following a chapter on objectivity, the authors discuss the role of reliability and validity and the problems that arise when these issues are neglected. They present a paradigm for the qualitative research process that makes it possible to pursue validity without neglecting reliability.