"Two monologues do not make a dialogue." - Jeff Daly

Focusing on evaluating speaking and listening, I think we all understand the pressure most students feel when it comes to take a test. All these feelings affect their performance most of the time, which would make the outcomes lack from the necessary reliability to evaluate their performance. Thus, it is important to consider constant assessment based on observation of students' performance during classes, regardless scoring, in order to have a clearer idea of their abilities.
To conclude with this entry, I'd like to superficially refer to speaking tasks which allow the students to performe with more freedom, such as sentence/dialogue completion tasks ( intensive speaking). Although there exists some predictable output, as speaking is one of the most flexible skills (if not the most flexible one), there is always room for diverse responses. In my opinion, these responses should be considered part of the students' oral production even when the content is not the expected one. That is why it could result more accurate to assess those responses taking into account standard traits such as pronunciation, fluency and vocabulary use, in order to evaluate students' performance in a more equal way.