IATEFL 2013 - On Listening Tasks and Tests

Attending and presenting at both TESOL and IATEFL conferences was a rewarding experience.  I always have two different perspectives when I attend and when I conduct a workshop. Attending a conference is a moment in which you see new trends in language teaching. We have contact with different and often revisited  viewpoints of what we sometimes  believe are unchangeable truths, and we have the priceless opportunity to meet old and new friends, professionals  who have a lot to share with you. As a presenter, I feel that a conference is a moment for networking and assessing the repercussion of the material you have been developing. Both are very motivating and make us want to share and learn even more. It is a never ending endeavor. I am sharing here an enriching presentation that I attended at IATEFL Conference in Liverpool, 2013 - Listening tests and tasks versus listening in the real world – by John Field (Oxford University Press). The talk outlined the types of mental processes involved in listening. Then it evaluated whether recorded material, formats, and items of conventional second/foreign language tests really tapped into this processes. Finally, suggestions were made for new forms of teacher-designed test and task that are more closely linked to real-world communication needs and to the listening construct.

Listening is a process taking place in the mind of the listener. The only way we can test the skill – or check understanding in the classroom – is indirectly - by asking questions. ELT teachers have to ask questions for three reasons: to test, to check understanding and to diagnose listening problems. This already distances the behavior of a learner or test candidate from that of a real-world listener. Then, what does a language test actually test?

We know that it is crucial for the learning process to consistently develop and assess the listening skill. We must, therefore, have in mind that it is impossible for a test to replicate the circumstances of real-life language use, but it is reasonable to ask to what extent a test (directly or indirectly) elicits from test takers’ mental processes like those that they would use in a real-world situation. This is a critical question in tests that claim to predict how well a candidate will perform in a real-world context, such as an academic institution, a professional position or an immigrant situation.

Cognitive validity is a well-established idea and educational researchers in the U.S. have investigated and questioned the following aspects of testing. Does a test of physics show that the learner can think like a physicist? Does a test of logical thinking test what it claims to test? Does a test in Medicine just show that learners have mastered facts – or does it show that they have the ability to diagnose? These intriguing questions lead us to reflect upon what listening consists of.

According to Mr. Fields, the model of expert listening starts with a speech signal – decoding and word search – and is followed by word parsing – separating the sentences into grammatical parts, such as subject, verb, etc. – which eventually leads to meaning construction. This model may question whether present listening tests / listening tasks materials elicit behavior from the listener that is like real-world listening processes, if they are comprehensive enough to cover most or all of the processes involved in listening, and if they are graded in a way that reflects learners’ development as listeners. He concluded that listening tests / tasks materials provide listeners with scripted (or even semi-scripted) recordings with little resemblance to natural everyday English, actors who mark commas and full stops, lack of hesitations and false starts, quite long utterances and regular rhythm, and voices that do not overlap. Aside from that, test setters sometimes put in distractors, making the recording much more informationally dense than a natural piece of speech would be.

The difficulty lies in the recording itself. Test designers and teachers tend to judge the difficulty of a piece of listening and even what points of the information to focus on by referring to a taspescript. However, these decisions also need to be made when listening to the recording. What parts of the recording (words or points of information) are prominent and easy to recognize? What characteristics of the speakers might make the recording more difficult? To choose recorded materials, teachers  have to take into consideration if it is authentic, recorded, scripted or improvised, analyze how now naturally the speakers include hesitations, for example, how fast they speak, how precisely the speakers form their words, the degree of formality, accents, if it is a dialog/conversation/interview, the frequency of the vocabulary uses, the complexity of grammar, the familiarity with the topic, the length of the recording, how dense the idea units are in the recording, how clearly structured is the overall line of argument and how concrete or abstract are the points made.

Mr. Fields concluded by affirming that conventional formats – multiple choices, gap filling, visual matching, true/false, multiple matching, identifying the speaker who said - require the listener to map from written information to spoken, eliminate negative possibilities as well as identify positive ones (multiple choices and True or False), read and write as well as listen (gap filling), and engage in complex logistical tasks which take us well beyond listening (multiple matching). He also claims that lower level learners understand far less than we assume, listen out for prominent words and try to match them to words in their vocabulary, are dependent on picking up salient words rather than chunks and whole utterances, a tendency that is increased by the use of gap filling tasks that only focus attention on word level.

He finally suggested that we provide items after a first playing of the recording and before a second. This ensures more natural listening without preconceptions or advance information other than the general context.  He insisted that we keep items short, since loading difficulty on to items just biases the test in favor of reading rather than listening. He made sure we use tasks that allow the test setter to ignore the order of the recording and to focus on global meaning rather than local detail. The information provided by Mr. Fields may not be new to many of us, but it always wonderful to listen to a specialist confirm or deny our assumptions, basing his conclusions on accurate research and studies. That is why attending a conference can make a difference in our lives.