Reflection 4

Norm-referenced standardized test score VS criterion-referenced test scores

I think no matter a teacher develops a lesson or a test, it is important to bear in mind that purpose is always a matter. Having a clear purpose, teachers will have a basic understanding of what he needs for assessing students. This week lecture introduces two terms “norm-referenced standardized test score” or “criterion-referenced test scores”. It’s sad to say sometimes, “norm-referenced standardized test score” is commonly adopted by most of the universities, in which some capable students may not achieve what they deserve as there are set target numbers in each grade level. Norm-referenced standardized test (NRT) score system does not necessarily mean less cooperation and collaboration between one another, and a less friendly study atmosphere. According to me personal experience, every student always strives for their best to achieve higher score, cooperate with school mates even more closely,  brainstorm ideas together, and give excellent academic performance as this system encourages the best to strive for the best. I do understand the competitiveness and competence of students may directly result in a good reputation of a tertiary institute.

Of course, criterion-referenced test scores (CRT) also suitable in some situation (but it also depends on what the test is going to be used for). I was a coordinator in my mother school when I first became an alumnus. I helped my school in new student intake process and had the experience of setting a test with my teachers. We used CRT for the entrance test (admission) to measure capacity or potential to learn in a level, in which the student’s abilities are most likely above average to study in our school. In this case, CRT is more suitable than NRT. All in all, purpose is always an important issue to think about before deciding which test score system to be used.

 

Reflection 3

I really like SBA

Under the new 334 education reform, it introduces a kind of new practice of school based assessment (SBA) of diploma of secondary examination (DSE). The introduction of SBA encourages formative assessment (still DSE is seen as a kind of summative assessment, which measures students’ academic performance) in the Hong Kong assessment system.

I remember I like SBA the most when I studied in secondary school. I often shared my book with my friends and we talked a lot about that book during lunch time. My teacher was very straight to give us score and not many of us could achieve “A” level in our performance. I read the book, memorized my lines (which I put in a notecard), rehearsed few times before assessment, have peers corrected my grammar… I got very well-prepared and I was confident in delivering my speech and sharing my book. I found myself very much like that type of assessment. I was assessed in a different way, not just to sit in a classroom with paper and pen. The process allows me to make myself independent of the teacher and practice self-learning. I like SBA (from a student’s perspective) because individual students have individual needs and problems in learning. I find SBA is more self-paced (I schedule all my work: when to finish reading the book, when to write the review, and when to proof-read my draft) and self-accessed (I often corrected my grammar and changed the words  to enhance message delivery) in a way.

Reflection 2

How to define a “good test”?

Tests come in many different shapes and forms. They have different purposes. Before discussing what a good test is, let us ask ourselves why we want good tests? There are many reasons… Tests act as the guarantor for the production system that guarantees that it delivers the value the customer is paying for. It is therefore important to remember that the tests are at least as important as the production code. A good test can be defined with the consideration of validity; reliability; practicality; wash back; authenticity and equity.

A good test is valid and reliable, but a valid and reliable test doesn't necessarily mean a good test. It is really hard to set or develop a good test with all the consideration mentioned above. For example, a test can be reliable not valid when the items that students can choose in MC questions are objective enough but students may use deduction as exam skill to eliminate answers; besides,  a test can be valid but not reliable when examiners and topics are different such as speaking and listening tests. In short, reliability and validity are important in good test development as reliability is a measure of the test's consistency, while validity is a measure of usefulness.

Reflection 1

Assessment = examination / test?'

I am always scared when I do my exams. But sadly to say, tests seem to be an unavoidable thing in teaching and learning. In my concept, test and assessment are not two “very” different terms in education and they are always seen as the same thing. In fact, test is only a genre of assessment, and assessment is an ongoing process that encompasses a wide range of methodological techniques, but not a one-time evaluation/measurement of students’ language/academic skills. These types of assessments: entrance test, placement test and diagnostic test can somehow be used as database for developing future teaching plans to facilitate students’ learning. To be brief, there are few important steps of developing a test: starting from purposes, defining the domain, specifying indicators of targets, writing up items, and leveling etc.

In language assessment it converts abstract concepts and constructs into observable language behaviors. The new 334 educational reform, DSE, is one of the examples of achievement test that measure and determine students’ achievement in terms of interpersonal skills, knowledge, language skills and experience.