This report addresses the reliability of the American Council on the Teaching of Foreign Languages (ACTFL) Oral Proficiency Interview (OPI), not as a measure of speaking ability, but rather as practiced by testers trained by the ACTFL, such as by the Interagency Language Roundtable (ILR), in English as a Second Language (ESL), French, German, Russian, and Spanish. Inter-rater consistency was measured by Pearson product-moment correlation coefficients and by a modified Cohen's kappa. Pearson coefficients were highly significant and remarkably similar in all five languages; Cohen's kappa results were also significant. Study results also confirm that interaction with the interviewee presents a source of variance in the assessment of speaking ability and that some levels of speech performance are simply harder to rate than others. Findings suggest that similarities and differences existed in the five languages that were difficult to explain and that inter-rater disagreement was very frequent and dependent on the level. It is concluded that a large and heterogeneous group of ACTFL-trained oral proficiency interviewers can apply the OPI in the five languages tested with a fairly high degree of consistency. (Contains 20 references.) (NAV)