PRINCIPLES OF LANGUAGE ASSESSMENT:
VALIDITY
Definition:
Tests for determining whether a measure is measuring the concept that the researcher thinks is being measured
A valid test:
- measures exactly what it proposes to measure
- does not measure irrelevant or “contaminating” variables
- relies as much as possible on empirical evidence (performance)
- involves performance that samples the test’s criterion (objective)
- offers useful, meaningful information about a test taker’s ability
- is supported by a theoretical rationale or argument
Face validity
Validity determined impressionistically, such as by asking students if the exam aligns with their expectations.
Tests should resemble test expectations at first impression.
If students perceive questions as not part of a test, the test may not be valid. This could lead to students not seriously answering the questions, affecting the test's validity.
Construct validity
Refers to the validity of the theoretical constructs the test measures.
Considered by some authors as the most critical type of validity.
Reflects if a test measures what it's supposed to
Concurrent validity
Content validity
Predictive validity
Uses a reputable test to validate one's own test.
Compares student performance on the new test with the reputable test
A correlation coefficient is obtained, with a closer correlation indicating higher concurrent validity.
Predictive validity is closely related to concurrent validity in that it too generates a numerical value.
refers to the representativeness and comprehensiveness of a test's content.
It's crucial to ensure the test accurately measures its intended purpose. For instance, a language course exam might not be valid if it only covers one language element.
In a conversational course, the final exam might not be valid if it requires demonstrating social skills.