Previous researchers have had difficulty in demonstrating relationships between dependence and competence in children and parental variables. This study involved a re-examination of this important question as well as a reconceptualization of the variables. Certain novel (in addition to some well worked) hypotheses were tested in a more pragmatic methodological framework. The novel methodological features of the study emanated from a concept of trait derived from a variant of ordinary language philosophy by Ossorio (1966). The methodological features included: (1) an emphasis upon direct observation of all variables (which were compared to sundry paper and pencil instruments), (2) a reconsideration of the necessity of obtaining "satisfactory" interjudge reliabilities and its relation to validity, and (3) the exploration of the extent to which different judges employ differing concepts corresponding to the same trait names.

From eleven nursery preschools 116 four-year-old boys were observed for six months. From this number 34 of the clearest examples of competent and dependent children were selected for further study in the laboratory. The parents of these children were next observed in interaction with their sons. The interaction was observed and rated by three judges while they completed various paper and pencil instruments during the first hour of their participation. During the second hour the parents and child were interviewed in three separate rooms.

It was found that the parents of competent children (when compared to parents of dependent children) treated their sons more as a child and less as an adult. Treating the child as a baby did not distinguish the two groups of parents. Also, parents of competent children were judged to be significantly more permissive and less restrictive, warmer and less hostile, more competent (as models for their child), and more consistent in philosophy and action (nature) than were the parents of dependent children. Hypotheses concerning the infantilization, and the dependence of the parents were not supported by the findings from this study.

Other findings included indications that divorce was more frequent among the parents of competent than dependent children and that parents (especially mothers) of competent children spent less (although possibly "better") time with their sons casting doubt on some studies reporting the contrary.

The correspondence among several levels of global judgments for the dependence and competence of the child and the matrix of specific elemental or atomistic interactions among the children were analyzed.

It was noted that although the interjudge reliabilities of certain of the assessment techniques were not particularly high, the findings were nonetheless significant for these variables. The implications for this and the other results of this study were discussed. It was suggested that the successes of the pragmatic methodology employed in this study might warrant researching other parent-child relationships believed to have been settled by findings of no significance. [148 pp.]