Wednesday, May 21, 2014

The Big Bad Monster of Assessment and How to Tame It by Maria Sachpazian - Report

In this workshop we examined the principles of well-rounded assessment and Ms Sachpazian gave us tips on how to design original tests.

First question: Can somebody learn without any kind of assessment?
The answer is that everything in life (not only at school) is somehow being assessed.

What is the difference between assessment and testing?
Assessment is much wider than testing.

Testing is associated with: exams, anxiety and… parents.
Unfortunately, tests don’t stick to the general abilities of our students. 
Ms Sachpazian compared tests to X-rays or blood tests. Tests show part of what the student can do.

So, what kinds of tests are there?
  • Placement tests
  • Progress tests
  • Achievement tests
  • Public examinations
  • Continuous Assessment.

When creating a test, we have to think about…
  • Is the test planned?
  • What do we test?
  • Are tests balanced?
  • Should teaching resemble testing or should testing imitate teaching?
  • Who designs the test?


Then we talked about validity.
Does the test test what it sets out to test?
There are different kinds of validity:
  • Face validity (does the test look valid?)
  • Content validity
  • Results Validity (i.g. AΣΕΠ exams)
  • Productive validity (mock tests)

Reliability is also important. There have to be consistent results whenever a test is administered.
Marker reliability is also essential. What affects the marker most is knowing the person who is being examined.
So, it is crucial to weigh the test before marking it, to create a marking scheme and to create a key in order to limit ambiguity.

Ambiguity has to do with unclear instructions and it is open to interpretation. Allowing too much freedom means less reliability and less standardization

A test should also be practical. We have to think:
How cost effective is the test? How much time will the students spend taking the test and how long does a teacher need to mark them?
How much data does the test reveal?
Does it contain discreet items or integrative language? Direct or indirect items?

To what extent does the language tested reflect authentic language?

We were then given suggestions about how to create a “good test”, such as collaborate with other teachers for new ideas or re-use materials from the book to make the test more familiar to the students.

All in all, it was a very thought provoking workshop which left us with a lot of material to work on.



By Theodora Papapanagiotou

No comments:

Post a Comment