Levels of measurement
This scale of measurement classifies data into distinct categories that can only be be assigned to one group
What is nominal?
This term refers to accuracy and consistency
What is reliability?
We repeat a test with the same client who hasn't had a change (expect it to be the same or similar so that it is reliable overtime)
What is test-retest?
The degree to which the scores on the assessment reflect the gold standard. This is really only used when there is already a gold standard assessment in place
What is concurrent validity?
This type of validity looks at whether all the items measure the same construct
What is Structural Validity?
This scale of measurement is the highest level of measurement, like interval, but there IS an absolute zero
What is Ratio?
This term relates to whether or not the assessment is measuring what we intend to measure. So are the scores providing us information about what it is intended for
What is validity?
When 1 assessor repeats the same test multiple times with a client and gets similar results
what is INTRA-rater?
This is when no gold standard exists
What is construct validity?
The minimum amount of change in a patient score that assures that the change isnt the result of error in measurement
What is minimal detectable change (MDC)?
This scale of measurement ranks data, space between IS NOT known or consistent
What is Ordinal?
This term refers to how well the assessment tool detects change in the person. So if the person is changing we want that assessment to be ____ to that and pick up on those differences.
What is responsiveness?
When we do an assessment and 2 different people can provide that assessment to 1 client and get the same or similar results
What is INTER- rater?
When a tool is translated for use with a different culture it looks at how well this new version reflects the performance of the original tool
what is Cross-cultural validity?
How well do the scores on this new assessment relate to the other similar measures and is the relationship aligned with the theory that supported the development of that assessment
What is hypothesis testing?
This scale of measurement is similar to ordinal; however the space between IS known, no absolute zero
This term refers to how useful the information. Just because an assessment is statistically significant does not always mean that it us useful or meaningful.
What is interpretability?
Inherent in all of the assessments that we do that can relate to the client (motivation), or our own skills (the training we've had on the assessment) and can be impacted by the environment (noise of testing space)
What is measurement error?
How well does the new assessment predict the gold standard outcome that we're looking for
What is predictive validity?
Smallest amount of change in an outcome that might be considered important to either the clinician or the client
What is minimally clinically important difference (MCID)?
This term describes how well a measure correlates with other measures of similar constructs.
What is convergent validity?
This concept is shown when a new balance assessment correctly identifies fall risk in older adults, matching real-world outcomes.
What is validity?
Refers to how well different items on one assessment relate to each other
What is internal consistency?
Looks at whether the assessment appears to be testing what its intending to test
What is Face Validity?
If we test the same person multiple times, we measure how varied are the scores
What is standard error of measurement (SEM)?