Stires's (1982) critique of our study (Levine, McDonald, O'Neal, & Garwood, 1980) and Knowles's (1982) comment on the state of classroom research (both in this issue) are considered. Stires's statement that we interpreted our stud)' as supporting the self-selection rather than the environmental hypothesis is shown to be inaccurate. Instead, we contend that some variables are affected more by location, others by self-selection, and still others by an interaction between the two. Stires's (1980) results are reevaluated and found to be congruent With this View'. We found Knowles's discussion of Griffith's (1921) study and his comments on classroom research provocative, although the relevance to our study seems tangential We fully agree with Knowles about the paucity of theory in this area and suggest several limitations in the existing research that should be dealt with systematically within a theoretical framework.
Read full abstract