CAL: Topics: Dialects
 
CAL Resources Archive
The CAL Resources Archive was created to provide our visitors with access to older pages and content from our Web site that they may find useful. Please be aware that information within the CAL Resources Archive is historical in nature and will not be maintained or updated by CAL.
 

ERIC Identifier: ED317036
Publication Date: 1989-12-00
Author: Stansfield, Charles W.
Source: ERIC Clearinghouse on Languages and Linguistics Washington DC.

Simulated Oral Proficiency Interviews

This "Digest" assumes general familiarity with the Oral Proficiency Interview (OPI) and with the Interagency Language Roundtable and the American Council on the Teaching of Foreign Languages skill level descriptions. For more information on the OPI, see the "ERIC Q & A, Testing Speaking Proficiency: The Oral Interview," by Pardee Lowe, Jr. and Judith E. Liskin-Gasparro. (ERIC Document Reproduction Service No. ED 232 483).

The simulated oral proficiency interview (SOPI) is a type of semi-direct speaking test that models, as closely as is practical, the format of the oral proficiency interview (OPI). The OPI is used by government agencies belonging to the Interagency Language Roundtable (ILR) and by the American Council on the Teaching of Foreign Languages (ACTFL) to assess general speaking proficiency in a second language (Liskin-Gasparro, 1987).

The SOPI prototype is a tape-recorded test consisting of six parts. It begins with simple personal background questions posed on the tape in a simulated initial encounter with a native speaker of the target language. During a brief pause, the examinee records a short answer to each question. Part one is analogous to the "warm-up" phase of the OPI. The remaining five parts are designed to elicit language that is similar to that which would be elicited during the level check and probe phases of the OPI. In order to avoid testing listening or reading ability, the remaining stimuli are in English. Parts two, three, and four employ pictures in a test booklet to check for the examinee's ability to perform the various functions that characterize the Intermediate and Advanced levels of the ACTFL proficiency guidelines, or levels one and two of the ILR skill level descriptions. Thus, the examinee is asked to give directions to someone using a map, to describe a particular place based on a drawing, and to narrate a sequence of events in the present, past, and future using drawings in the test booklet as a guide. Parts five and six of the SOPI require the examinee to tailor his or her discourse strategies to selected topics and real-life situations. These parts assess the examinee's ability to handle the functions and content that characterize the Advanced and Superior levels of the ACTFL guidelines, or levels two through four of the ILR skill level descriptions. Like the OPI, the SOPI ends with a "wind-down." This is usually an easy question designed to put the examinee at ease and to facilitate the ending of the examination in as natural a manner as possible.

After the test is completed, the tape is scored by a trained rater using a combined ACTFL/ILR scale. Scores may range from the Novice level to High Superior. The latter score is equivalent to a rating between 3+ and 5 on the ILR scale.

As indicated above, the SOPI is a type of semi-direct test. Clark (1979) defined a semi-direct test as one that elicits speech by means of tape recordings, printed test booklets, or other non-human elicitation procedures. A semi-direct test can employ a wide variety of item formats. These may include techniques such as spoken pattern practice in response to cues in the test booklet or on tape, reading aloud, sentence repetition, sentence completion, naming nouns or verbs depicted through line drawings in the test booklet, describing a single picture or describing a picture sequence (Clark, 1979; Clark & Swinton, 1979). Many of these semi-direct elicitation techniques are inherently different from the relatively authentic, context-based techniques that would be found in the OPI and in the SOPI.

RESEARCH ON THE SOPI

In five studies involving different test development teams and different languages, the SOPI has shown itself to be a valid and reliable surrogate of the OPI. Clark and Li (1986) reported on the development of four forms of a simulated oral proficiency interview in Chinese, which were then administered, together with an OPI, to 32 students of Chinese at two universities. Each test was scored by two raters and the scores on the two test types were statistically compared. The results showed the correlation between the SOPI and the OPI to be .93.

Stansfield and Kenyon (1988) reported on the development of three forms of another SOPI called the Portuguese Speaking Test. This test and an OPI were administered to 30 adult learners of Portuguese at four institutions. Each test was scored by two raters. In this study, a correlation of .93 between the two types of test was also found. In addition, the SOPI showed itself to be slightly more reliable and easier to rate than the OPI.

Shohamy et al. (In Press) reported on a joint project between the Center for Applied Linguistics and the University of Tel Aviv that developed and validated another SOPI, the Hebrew Speaking Test. Two forms of the test were developed for use at Hebrew language schools for immigrants to Israel, and two forms were developed for use in North America. The first two forms were administered to 20 foreign students at the University of Tel Aviv and the other two forms were administered to 20 students of Hebrew at two U.S. universities. Each group also participated in an OPI. The correlation between the OPI and the Israeli version of the SOPI was .89, while the correlation for the U.S. version was .94.

Most recently, Stansfield and Kenyon (1989) reported on the development and validation of SOPIs in Indonesian and Hausa. In the Indonesian study, the correlation with the OPI for 16 adult learners was .95. Because no ACTFL or ILR-certified interviewer/raters were available for Hausa, it was not possible to administer an OPI to the 13 subjects who took the Hausa Speaking Test. However, two Hausa speakers were trained in the ACTFL/ILR scale and they subsequently scored the test tapes on that scale. The raters showed high interrater reliability (.91) in scoring the test and indicated that they believed it elicited an adequate sample of language from which to assign a rating.

THE SOPI VS. THE OPI

In comparison with the OPI, the SOPI would seem to offer certain advantages. The OPI must be administered by a trained interviewer, whereas any teacher, aide, or language lab technician can administer the SOPI. This may be especially useful in locations where a trained interviewer is not available. The SOPI can be simultaneously administered to a group of examinees by a single administrator, whereas the OPI must be individually administered. Thus, the SOPI may be preferable when many examinees need to be tested within a short span of time.

In addition to these practical advantages, the SOPI may offer psychometric advantages in terms of validity and reliability. Although the OPI varies in length, it typically takes 20 to 25 minutes to administer and produces 12-15 minutes of examinee speech. The SOPI takes 45 minutes to administer and produces a longer sample, usually 20-23 minutes of examinee speech. The more extensive sample may contribute to a more valid assessment.

In an OPI, the validity of the test sample elicited is largely determined by the skill of the interviewer. Interviewers can vary considerably in their interviewing techniques, yet the SOPI offers the same quality of interview to each examinee.

The SOPI also helps ensure high reliability. By recording the test for later scoring, it is possible to ensure that examinees will be rated by the most reliable raters. In the OPI, the same individual typically conducts the interview and scores the test. Yet the interviewer may not be the most reliable or accurate rater. Also, raters who have scored both types of test have reported that it is often easier to assign a rating to a SOPI performance. In part, this may be because the SOPI produces a longer speech sample and because each examinee is given the same questions. Thus, distinctions in proficiency may appear more salient to the rater.

CONCLUSION

An examination of the SOPI research, which has been carried out on different subjects and on tests of different languages produced by different test development teams, shows that the SOPI correlates so highly with the OPI that it seems safe to say that the tests measure the same abilities. Also, a comparison of the advantages of each suggests that the SOPI offers certain practical and psychometric advantages over the OPI. Thus, it may be useful to consider the circumstances that should motivate the selection of one format or the other.

Since the tasks on the SOPI are ones that can only be effectively handled by responding in sentences and connected discourse, the SOPI is not appropriate for learners below the level of Intermediate Low. Similarly, the semi-direct format of the test does not permit the extensive probing that may be necessary to distinguish between the highest levels of proficiency on the ILR scale, such as levels 4, 4+, and 5.

The purpose of testing may also play a role in the selection. If the test is to have important consequences or is to be used for research, it may be preferable to administer a SOPI, since it provides control over reliability and validity of the score. Such a situation might be found in the use of a proficiency score to determine whether or not applicants are qualified for employment, such as for teacher certification purposes. On the other hand, if scores are to be used for placement or diagnosis within an instructional program and a competent interviewer is available, it would seem preferable to administer an OPI. In such a situation, an error in placement can be easily corrected. Similarly, an OPI administered by a competent interviewer may sometimes be preferable for program evaluation purposes because of the qualitative information it can provide and because the score will not have important repercussions for the examinee.

REFERENCES

Clark, J.L.D. (1979). Direct vs. semi-direct tests of speaking ability. In E.J. Briere & F.B. Hinofotis (Eds.), "Concepts in language testing: Some recent studies"(pp.35-49). Washington, DC: TESOL.

Clark, J.L.D. & Li, Y. (1986). "Development, validation, and dissemination of a proficiency-based test of speaking ability in Chinese and an associated assessment model for other less commonly taught languages." Washington, DC: Center for Applied Linguistics. (ERIC Document Reproduction Service No. ED 278 264)

Clark, J.L.D. & Swinton, S.S. (1979). "Exploration of speaking proficiency measures in the TOEFL context (TOEFL Research Report 4)." Princeton, NJ: Educational Testing Service.

Liskin-Gasparro, J. (1987). "Testing and teaching for oral proficiency." Boston, MA: Heinle and Heinle Publishers.

Shohamy, E., Gordon, C., Kenyon, D.M., & Stansfield, C.W. (In Press). The development and validation of a semi-direct test for assessing oral proficiency in Hebrew. "Bulletin Of Hebrew Higher Education," 4 (1).

Stansfield, C.W. & Kenyon, D.M. (1988). "Development of the Portuguese speaking test." Washington, DC: Center for Applied Linguistics. (ERIC Document Reproduction Service No. ED 296 586)

Stansfield, C.W. & Kenyon, D.M. (1989). "Development of the Hausa, Hebrew, and Indonesian speaking tests." Washington, DC: Center for Applied Linguistics. (ERIC Document Reproduction Service, forthcoming)

 

Untitled Document
CAL Home | CAL Store | Press Room | Jobs | Contact Us | Site Map | Privacy | Links
Copyright © 2014 CAL