The frequently referred-to "generation gap" is a myth as yet unsubstantiated by empirical evidence. Explicit definitions of the phenomenon are lacking; implicit definitions are ambiguous and hard to extract. To offer an empirically tested operational definition of "generation gap" and analyze its relationship to several independent social variables, a questionnaire was given to the freshman and junior classes of each of 4 high schools in the midwest, and on the west and east coasts. Using as the definition of generation gap "index of identification with Adult Authority Figures" (i.e., parents, teachers, administrators, policemen), considerable variation was found. A steady increase in adult identification was found with increase in socioeconomic status. Success in school, as measured by grades, and identification with certain traditional middle-class ideals, for example gratification deferral, could also be used to predict adult identification. Religion was also a factor--Jews proved most likely to identify with adult authorities, and unchurched and Liberal Protestants, least likely. Age and sex, however, made no appreciable difference in identification with adult authorities. (KS)