Survey, interview and desk research indicate that universities tend to fund a PELA through central departments. Some, however, are funded by individual faculties. In a very few cases, the introduction of an institution-wide PELA was funded through special funds. In no case do students have to pay directly. Not all universities with PELAs have developed their own assessment, although this is by far the most common model. The different models include:

  • In-house design (either using existing staff or employing a consultant).
  • Commercially available product, sometimes adapted to the needs of a specific institution.
  • Collaboratively developed design, usually across universities.

Whichever approach is taken, it is essential to commence with the clear understanding that a large-scale PELA is not a cheap option if it is to be a genuinely educational tool. There are upfront costs associated with design of assessment items, piloting of a new instrument, analysis of results, review, implementation and continuous updating of content. There are also costs associated with administration (e.g. venue, supervision, computing equipment, ICT administration) and with marking unless the process is automated. There are also costs of disseminating results, taking action when the results require it, and monitoring the processes involved. Finally, there are costs involved in training staff to administer, mark and interpret the results of the PELA. These costs assume that an institution has available staff with language testing and assessment expertise on hand. If this is not the case, then it is unlikely that an in-house design will provide a credible instrument, and institutions are likely to be better off seeking external assistance.