In this study four research strategies used in previous studies of the socialization of cognitive skills were first contrasted and then assessed in terms of five criteria. (The data base for this study was 32 videotapes of mothers preparing their 7- or 9-year-old children to take a memory test.) The four strategies (frequency coding, fine-grained sequential analysis, single-case ethnography, and comparative ethnography) are discussed in terms of their utility for providing an empirically adequate account of how individuals achieve meaningful communication within the context of an evolving interaction. Strategies are then evaluated in terms of (1) the questions each method can answer, (2) procedures related to data analysis, (3) problems of assessing the validity of the analysis, (4) problems of assessing the reliability and validity of data collection, and (5) limits each strategy imposes on statements concerning the generality of the findings. While all of the strategies were found to have utility for examining particular aspects of data on the socialization of cognitive skills, neither frequency coding nor fine-grained sequential analysis were found to represent adequately the complexity of the dyadic interaction. The single-case ethnographic method, however, was found to provide accounts of how participants jointly accomplished cognitive tasks, how mothers established and maintained a common framework, and how both mother and child dyads worked toward a transfer of responsibility for the direction of the task. In order to examine the prevalence of structures of interaction found in the single-case ethnographies, two multi-case, comparative ethnographic studies are presently being conducted. (Author/RH)