Measurement Flashcards
What is validity?
The degree to which the test measures what it is designed to measure.
What is reliability?
The consistency or repeatability of test scores from trial to trial or day to day.
What is objectivity?
The consistency of scores obtained by MORE THAN ONE TESTER
What is another name for objectivity?
Rater reliability
What does it mean if a test has HIGH objectivity? Give an example.
It should not matter who scores the test (the results would be the same). Ex: Multiple Choice
What does it mean if a test has LOW objectivity? Give an example.
The score depends on who is scoring. Ex: Any sport where you are GIVEN a score… you do not make your score.
True or False?
For a test to be valid, it must also be reliable.
False. For a test to be valid, it must also be reliable and objective.
True or False?
A reliable test is not always valid.
True.
What are the types of validity?
- Logical Validity
- Content Validity
- Criterion Validity
- Construct Validity
What is logical/face validity?
Does the test look like it measures what it is supposed to measure on the SURFACE
What are the problems with logical/face validity?
- Not enough evidence to say a test is valid
- Not statistical
What is content validity?
Knowledge Test: The content must match what was covered in class to the proportions covered in class.
Psychological Test: The content must match the intent of the test
What types of tests are usually used for content validity?
Paper/pencil tests (knowledge and psychological tests)
What is the problem with content validity?
Typically, not statistical
What are the two types of criterion validity?
- Concurrent
- Predictive
Describe concurrent validity.
Comparing your test to a known valid test/the gold standard.
How is concurrent validity usually done? How do you know if your test is good?
With a correlation.
A good correlation is close to 1 or -1.
What has a strong correlation to the mile-run field test?
The CV test and VO2max
What has a strong correlation to the skinfold field test?
DEXA
Describe predictive validity.
How well a test can predict achievement in the future. Usually a set of measures, not a single measure.
Give 2 examples of predictive validity.
- Admission to upper education - likelihood of graduating
- NFL Draft - physical/psychological/cognitive tests
What is construct validity?
How well does the test measure the underlying construct you want to measure?
True or False?
Constructs are always visible.
False. Constructs may or may not be observable.
What is the MAIN idea of construct validity?
We are trying to validate something that is not observable.
When is construct validity usually seen?
In psychological tests.