“…Rather than testing Hypothesis a, many past studies have assumed its validity and thus have taken social class indices as a convenient, if somewhat rough, index of differences in the pattern and type of communicative interaction. Much of the evidence of social class differences in parent-child interaction has been subjective, nonquantitative, and/or based upon questionnaire or interview responses rather than observation of actual social interaction (e.g., Bossard, 1945;Brandis & Henderson, 1970;Bronfenbrenner, 1958;Higgins, 1973;Keller, 1962;Marge, 1965;Miller, 1958;Milner, 1951;Pavenstedt, 1965;Walters, Connor, & Zunich, 1964).…”