Examination of Data-processing Methods When Observing Interactions in Infancy and Early Childhood

幼儿互动观察的数据处理方法研究本研究由两部分组成。第一部分内容如下；首先，用录像机记录了六对母子的自然情境关系。第二，由两名观察员用不同的方法检查录像带。一种方法是用秒表记录观察结果，并在观察单上手写笔记。另一种是用个人电脑记录的观察结果，即按编码数字键。与我们的假设相反，手写笔记的数据量比电脑记录的数据量要大。我们认为非常密集的训练是必要的，以准确的计算机记录观察结果。第二部分内容如下；首先，对基于马尔可夫模型的理论框架进行了严格的检验，以构建一个更全面和上下文敏感的框架。其次，使用上述新框架对幼儿教育的情况进行了详细检查。讨论了这些分析的意义。

Examination of Data-processing Methods When Observing Interactions in Infancy and Early ChildhoodThis study consists of two parts. The first part is as follows; First, natural-situation relationships between mother and infant (six couples) were recorded by VTR. Second, the video tapes were checked by two observers using different methods. One method was that of recording observations by stop-watch and hand-written notes on observation sheets. The other was observations recorded by personal computer, i.e. pushing coded number keys. Contrary to our hypothesis, the numbers of data from hand-written notes were larger than those from computer records. We think very intensive training is necessary for accurate computer-recording of observations.The second part is as follows; First, theoretical frameworks based on the Markov model were critically examined to construct a more holistic and context-sensitive framework. Second, a situation of early childhood education was examined in detail, using the new framework above. Implications of these analyses were discussed.