| The Educational Testing Service (1982) claim that "a well-structured oral proficiency interview tests speaking ability in a real life context--a conversation" (p. 13) has been so pervasive that Lazaraton (1992) comments: "clearly it is in the context of conversation that an oral proficiency interview can or must be examined because almost every piece of literature written about it, mistaken or not, refers to its conversational nature" (p. 375).;This dissertation was conducted to provide answers to the following research questions: (1) What kind of speech event is the OPI? Is it more like everyday conversation, classroom interaction, or an interview? (2) Do native speakers (i.e, testers and non-testers) differ in their judgments of the nature of the OPI communicative speech event?;To obtain some answers to the research question 1, a discourse analysis of 35 OPIs was conducted. Frequency counts and chi-square inferential statistics were performed on the transcribed and coded OPI data. The results of the discourse analysis were then compared with the prototypical features of the three speech events selected for the study (i.e., conversation, interview, and classroom interaction) to determine and describe the most typical discourse and linguistic features of the OPI.;To obtain detailed information about testers' and non-testers' perceptions of the OPI and its components, a semantic differential study was conducted. A two-way analysis of variance was performed on the semantic differential data to determine how testers and non-testers perceive/judge the OPI as a speech event.;In sum, the discourse analysis provided detailed information about what participants actually do in the OPI and the semantic differential provided detailed information on how people (i.e., testers and non-testers) perceive/judge what has been done in the OPI.;The findings of the discourse analysis and the semantic differential contradict the ETS claim. The OPI does NOT test speaking ability in a real life context--conversation. It tests speaking ability in the context of an interview, or more precisely, in a context of survey research interviews and sociolinguistic interviews. These findings raise a question about the validity of the OPI testing instrument. |
