RTT Retelling Method: An Alternative Approach to Intelligibility Testing

silewp2007_006.pdf194.89 KB
Statement of Responsibility:
Kluge, Angela
Series Issue:
Part Of Series:
SIL Electronic Working Papers 2007-006
14 pages

This paper presents an alternative approach to intelligibility testing, the RTT (Recorded Text Testing) retelling method. This method constitutes a modified version of the standard RTT method as described by Casad (1974) The standard method uses a short text recorded from an Li speaker of the speech variety being tested with questions about the text interspersed in appropriate places throughout the text. Respondents are required to answer these questions, which are in their first language (L1). This approach is based on the assumption that from the level of correctness that respondents have of the speech variety being tested. Subsequently, conclusions are drawn concerning degrees of interdialectal intelligibility and which variety or varieties of a given language area, network or continuum might serve best as the write reference dialect. In spite of its many strengths, experience over the years has indicated a number of difficulties with this method. Its question format is often found to be culturally inappropriate, requires indirect inference, and relies on difficult question selection. To overcome these difficulties, the RTT retelling method has been developed. This modified version of the standard RTT method requires respondents to listen to a recorded narrative. The narrative is broken down into natural chunks of one or two sentences each. Rather than hiving to answer specific comprehension questions, respondents are required to retell the recorded text segment in their L1.

Following a brief review of the standard RTT method and problems regarding its question format, this paper presents a detailed description of the RTT retelling method, including test design, and testing and scoring procedures, as well as a brief discussion of some of the advantages and disadvantages of the RTT retelling method.

Publication Status:
Subject Languages:
Content Language:
Work Type:
Intelligibility testing
Nature of Work:
Entry Number: