The verbal report is the experiencing person’s verbal description of an internal event such as a visual image or verbal thought. Verbal report can take many forms and like descriptions of external events can be more or less precise. Because observers vary widely in their ability and motivation to observe their own internal events, verbal reports cannot be taken as definitive evidence of the occurrence of the events they represent. This is especially true with young children, who are often unable to describe even the simplest and most obvious of their own internal events. Used with caution, however, verbal report, especially with older children, can serve as a helpful index of the occurrence and sometimes even of the nature of internal events.
Deduction from observables is employed when the researcher wishes to study internal processes that are unavailable to observation by anyone, even by the experiencing person. This is often the case in the study of cognitive development when processes of knowing constitute the subject matter for investigation. Indeed, studies of the development of what children know about objects, about other people (including other people’s thought processes), and about themselves constitute one of the most important areas of developmental research. To study such processes, researchers typically present the child with a variety of different but related tasks and observe the child’s pattern of activity (sometimes including verbal report) in response to the tasks. A set of rules that might describe the “knowledge” of someone who exhibits one rather than some other pattern of activity across the respective tasks is then constructed. The conclusion is then made that the workings of the cognitive system of the child who adopts such a response pattern conform to these rules. Further research is then typically carried out to assess predictions, based on this inference, about the child’s activity in other, related situations (Schore, 2000).
To continue, researchers must be able to represent the occurrence of a psychological event by some means that will be relatively permanent. This record, referred to as data, may be close to the original event (such as audio or video recordings) or relatively far removed from it (such as computerized storage of numbers read from a time clock). Whatever its relationship to the actual event, however, data must at least represent the occurrence of the event; usually it will represent one or another characteristic of the event as well.
This report describes the application of scientific observation methods to measure the reliability of coding using the method employed by Kogan and Carter (1996) in their ‘still-face’ experiment. It should be noted that these scholars were not the first ones to study the interactions between a mother and a baby.
