6 Assessing productive and interactive skills Personal refl ection

Input and interaction
For assessments of writing skills, the choice of response format 
generally lies between paper-based and computer-based options. 
Input might take the form of a written or visual prompt, or a 
combination of the two. In assessments of speaking, input is more 
often presented orally.
Figure 6.2
 
Interaction in the oral profi ciency interview (OPI)

---

 
Assessing productive and interactive skills 
139
In the classic oral interview (see 
Figure 6.2
), an examiner poses 
questions to the assessee either face to face or over the telephone. In 
effect, the examiner fi lls the role played in most tests by the question 
paper. However, individual assessees will encounter different examiners 
with different personalities and questioning styles.
The specifi c questions posed by the examiner and the manner in 
which they are put can affect the kind of performance that is elicited. 
It is therefore important to control questioning in a systematic way. In 
large-scale tests, attempts have been made to standardise examiner 
behaviour through training, by scripting or partially scripting their 
role, and by specifying other contextual features, such as what the 
interviewer should wear and where he or she should sit in relation to 
the assessee. This is intended to ensure that every assessee has a 
comparable experience (see Fulcher, 2003, 
Chapter 6
). 
An alternative to the face-to-face interview is to replace the examiner 
with a recording (see 
Figure 6.3
), but this brings its own problems. 
Recordings cannot interact realistically with the assessee, so turn 
taking and conversation management cannot be assessed. Some people 
feel very awkward talking to machines and may not perform at their 
best in these circumstances.
The interview format itself has come under criticism because the 
interaction it gives rise to tends to be very one sided. The interviewer 
asks all the questions and controls the interaction while the assessee is 
just required to answer. This does not closely resemble everyday 
conversations, discussions or transactions, where the participants have 
a more balanced relationship and share in directing the interaction. 
Some features of conversation that may be of interest to the assessment 
user play little or no part in an interview (Johnson, 2001).
A possible remedy is to employ paired or group speaking assessments 
(see 
Figure 6.4
). In this format, assessees are asked to interact with one 
Figure 6.3 
Interaction in the semi-direct or computer-based oral profi ciency 
interview (SOPI/COPI)
Figure 6.4
 
Interaction in the paired/group speaking assessment

---

140