Skip to Content

AMEP Research Centre

Development of the Assessment Task Bank

In the first phase of the Task Bank project, expressions of interest were sought from Adult Migrant English Program staff to form a national working group on assessment which would meet at the AMEP Research Centre at Macquarie University in Sydney. Once the working group was formed with representatives from NSW, ACT, Qld, SA, Vic and WA, teachers were again invited to submit samples of assessment tasks from Certificates II and III. In their professional Development workshops, the working group

  • vetted the tasks, selecting the most suitable ones for inclusion in the task bank and
  • modified them, drawing on the results of the research carried out in the context of the Special Projects outlined above.

For example, Gillian Wigglesworth had found that the role of the interlocutor was one of the most important factors in determining how well learners performed in speaking tasks. This information was used to rewrite the role cards for the speaking competencies, to ensure that the interlocutor role was clearly defined.

Once the tasks had been modified, the national working group participants piloted them with a small number of learners in their own centres. The purpose of the pilot was to:

  • detect any major faults in the tasks
  • make sure they produced performances that complied with the
    Performance Criteria of the Certificate in Spoken and Written English (CSWE)
  • check that the conditions of administration (corresponding to the range statements in the CSWE) were consistent
  • ensure that the tasks were of the right level of difficulty for the learners.

The national working group then met again and modified the tasks once more using the findings from the pilot study.

Finally, the tasks were trialled with large numbers of learners around Australia. In fact, many of you may have been involved in that trialling process. The trialling had similar objectives to the pilot study but with larger numbers to enable us to run statistical tests on the results. At this stage some of the tasks were rejected or modified once again and retrialled. Once the results of the trialling had been analysed, the tasks were calibrated. This means that they were ordered in terms of their difficulty. They were now finally ready to be put in the task bank.

This might help explain why the process of developing assessment tasks for a Task Bank has been such a lengthy one!