A NOTE ON THE RELIABILITY OF TWO METHODS OF PROTOCOL ANALYSIS.

For several months entering cadet classes were asked to write an answer to this question What do you think of a person who voluntarily attrites from the flight training program. Replies were sorted along the favorable-unfavorable continuum. Two psychologists found low consistency in the sorting of the protocols. A third psychologist and intensive efforts in developing detailed instructions for sorting did little to improve matters. It was concluded that the analysis of protocol responses to such a specified question as was asked, on even so simply a dimension as favorable-unfavorable, was a difficult task and deserved further consideration. The specific problem of protocol analysis dealt with in this study was the question of analysis by whole or gestalt analyses versus sentence by sentence or part analyses of protocols.