Douglas Altamiro Consolo

In this article I review principles and practical aspects of language assessment in foreign language (FL) learning, more specifically with a focus on Computer-Assisted Language Learning (CALL) and other contexts in which computers are used as a means for distance learning and for language assessment (Computer-Aided Assessment - CAA). Assessment constitutes an essential dimension of learning experiences and it is an aspect of most formal processes of language teaching and learning. Language assessment may involve language testing, as well as other procedures and instruments such as observations, performance tasks, portfolios and self-evaluation, and by combining information from various sources of assessment one is able to obtain more valid and reliable results. I draw on the literature on CALL and on language assessment, and on data collected within the scope of the Teletandem Brazil Project: foreign languages for all (henceforth TBP), to support my position on principles that may or may not characterize language assessment in the context of teletandem interactions. CAA is defined as any type of activity in which computers are used to support a process of assessment apart from and beyond their simple function to store and transmit information. CAA helps faster assessment, increases the quality and quantity of information detected and maximizes the provision of feedback about language assessment processes. In the TBP project, undergraduate students from a Brazilian university interacted with students from universities abroad, by means of computer programmes for synchronous communication, microphones and webcams. Besides the claims about CALL from the literature, I make reference to teletandem interactions in both EFL and Portuguese as a foreign language, considering occasions in which teletandem agents evaluate each other linguistic performances. I also analyse a questionnaire for evaluation in teletandem, which does not focus on language assessment but rather on the experience of interacting in the teletandem context and on the tools used for communication. No clear distinctions were found to exist between CAA and more traditional procedures for language assessment. Principles for CAA seem to combine traditional bases for language assessment and testing with a number of pedagogical principles that underpin distance learning. Nevertheless, CALL and CAA can contribute in various aspects of language education, especially when large numbers of learners are involved in teaching and learning processes. However, these principles do not characterize a new paradigm in language assessment, since the linguistic criteria on which teletandem agents base their evaluation are very similar to criteria that underpin language assessment and testing by means of paper-and-pencil tests, for example. I conclude the article indicating the need for further investigation and the establishment of principles for language assessment in electronic contexts.