Computational Analysis of Interactions
Brian MacWhinney · 1996
Child language research thrives on naturalistic data – data collected from spontaneous interactions in naturally occurring situations. However, the process of collecting, transcribing, and analyzing naturalistic data is extremely time consuming and often quite unreliable. One of the major methodological developments in the field of child language research over the past decade has been the introduction of computerized systems for dealing with the transcription, coding, and analysis of spontaneous production data. One such system – the Child Language Data Exchange System (CHILDES) – will be our focus in this chapter. We will begin by reviewing the background to the formation of the CHILDES system. Next, we will examine the basic CHILDES tools, and the ways these tools can be used to address particular research goals. The chapter will then explore a new range of analytic capabilities planned for the next few years.