Exploration of a quantitative method for measuring behaviors in conversation

Publication Type:
Journal Article
Citation:
Aphasiology, 2018, 32 (3), pp. 247 - 263
Issue Date:
2018-03-04
Full metadata record
© 2017 Informa UK Limited, trading as Taylor & Francis Group. Background: The literature on communication partner training (CPT) includes mainly studies with a small number of participants, because methods to measure changes in conversation pose practical challenges limiting the analysis of large samples. Aim: The aim of this study was to explore a quantitative procedure that would allow one to measure specific behavioral changes occurring in conversational exchanges involving a person with aphasia and a partner. Methods & Procedures: Forty-three problem-solving situations presented visually as well as with a simple written explanation were created to elicit conversation. In order to test the situations and develop further a procedure, we used data from a spouse of a man with aphasia during CPT delivered in a clinical setting. We developed specific definitions related to conversational behaviors targeted in the CPT. These defined behaviors were analyzed using a transcription-less method and an annotation software in the couple’s 39 conversation samples collected before, throughout, and 3-months post CPT. Reliability data were collected. Outcomes & Results: The procedure enabled us to create a protocol with two types of conversational situations and reliable definitions for measurement of conversational behaviors in a timely fashion. Pilot data of the measures are provided. Conclusions: It is expected that the method presented in this pilot study may be used to document the outcomes of CPT. It could be used with single-subject designs that require repeated measures and multiple group designs that require comparable data over large samples. It provides a method of data collection and analysis to better evaluate the effects of conversation-based treatments such as CPT.
Please use this identifier to cite or link to this item: