CHARACTERISTICS OF A GOOD TEST: VALIDITY

CHARACTERISTICS OF A GOOD TEST: VALIDITY (GROUP 6)

  Lies Nureni & Lukman Chamdani

A test is said to be valid if it is measures what it claims to measure.

Validity is a unified concept. As such, it has several dimensions or aspects, they are:

  • Face Validity

It relates to how a test looks to other people, students, experts, etc.

Example: a grammar test should test the grammar not the vocabulary. Thus, in a grammar test, the vocabulary should be easy and vice versa.

 

  • Content Validity

A test needs to have a representative sample of the teaching/instructional contents as defined and covered in the curriculum. How do we know that a test is valid in content? By using blueprint/kisi-kisi/rambu-rambu.

Example: a grammar test should contain every part of the grammar. But we cannot take all of them, so just take the sample to represent the content.

 

  • Empirical Validity

It relates to the closeness between the score obtained from a test with the other criteria outside that test. It is divided into two:

a)        Concurrent validity: how well the test estimates current performance on some valued measure other than the test itself.

e.g. TOEFL and TOEIC >> if someone’s TOEFL score is high, then we can say that automatically her/his TOEIC score is high too.

b)        Predictive Validity: how well the test predicts future performance on some valued measure other than the test itself.

e.g. GPA >> if someone’s GPA is high, we can say that s/he will have a high salary job.

 

  • Construct Validity

It relates to our understanding of the existing theory to construct a test.

Example: if we want to construct a speaking test, we have to find, compare, and criticize the theory about the speaking test not the other.

Construct validity requires two levels:

  1. Logical Analysis

ü  Dimension >> standard of competence

ü  Variable >> basic competences

ü  Sub variable

ü  Indicator

ü  Test items

 

  1. Empirical Analysis

Whether the test items measure what is defined by their indicators.

 

  • Washback Validity

It relates to the influence of a test to the teaching learning process. There are two types of washback:

  1. Positive Washback

Micro Level (classroom setting)

ü  Tests induce teachers to cover their subjects more thoroughly.

ü  Tests make students work harder.

ü  Tests encourage positive teaching-learning processes.

Macro Level (educational/societal system)

Decision makers (govt.) use the authority power of high stakes testing to achieve the goals of teaching and learning, such as the introduction of new textbooks and new curricula.

 

  1. Negative Washback

Micro Level (classroom setting)

ü  Tests encourage teachers to make “teaching to the test” curriculum.

ü  Tests bring anxiety to both teachers and students and distort their performance.

ü  Tests bring anxiety to both teachers and students and distort their performance.

ü  Tests drag students to learn discrete points of knowledge that are tested.

ü  Tests make students create a negative judgment toward tests, and alter their learning motivation.

Macro Level (educational/societal system)

Decision makers overwhelmingly use tests to promote their political agendas and seize influence and control of educational systems.

About these ads

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s