This paper documents the methodological study which was undertaken to discover the best way to obtain high quality tape-recorded samples of casual, spontaneous speech from children in kindergarten through Grade 3. All children interviewed were from schools in areas qualifying for compensatory education programs under Title I of the Elementary and Secondary Education Act of 1965. With the exception of two sessions, all group interviews were video taped. There were six groups of children, aged 5 to 9, participating in the interviews. The technique of interviewing children individually was tried out with Groups 1 and 2. Two paired interviews were conducted with children from Group 2 and three paired interviews were conducted with children from Group 4. The most important finding suggested by this survey is that it is possible to elicit and record casual speech samples from 5- to 9-year-old children in an artificial environment, provided great care is taken to make the children feel relaxed and unthreatened. Apparently, the best method for doing this is to choose one child from the socioeconomic group under study and allow him to form a group of peers at his discretion. (Author/CK)