Measurement Tools

Question 1

What are the types of measurement tools?

Answer 1

descriptive discriminative, predictive, evaluative and diagnostic

Question 2

What do descriptive measurement tools do?

Answer 2

describes the event/functional status

Question 3

What do discriminative measurement tools do?

Answer 3

discriminate between different concepts or qualities that we want to assess. Compare against normative data.

Question 4

What do predictive measurement tools do?

Answer 4

predict future events (may also discriminate between events). Can classify people into predetermined categories or can predict future abilities and outcomes

Question 5

What do diagnostic measurement tools do?

Answer 5

diagnose a health condition

Question 6

What do evaluative measurement tools do?

Answer 6

evaluate health outcomes, detect change over time. In order to do this they must be sensitive to change and should have a good test-retest reliability

Question 7

What is the purpose of measurement?

Answer 7

to give an operational/practical definition of the construct of the question/purpose

Question 8

What is the difference between 2 measurements called?

Answer 8

variation in the measurements/error

Question 9

When is error present?

Answer 9

always

Question 10

What influences error?

Answer 10

client’s attributes, evaluation/instrument, therapist/rater and environment

Question 11

What is the sum of measurement error?

Answer 11

observed score = true ability + random error +/- systematic error

Question 12

What is systematic error?

Answer 12

bias

Question 13

What is random error?

Answer 13

variability

Question 14

What is a reliable tool?

Answer 14

One that produces consistent results over repeated trials

Question 15

What can effect the stability of a tool?

Answer 15

translation of instruments - may not be culturally appropriate between different populations

Question 16

When would you use a measurement tool?

Answer 16

• to gather info
• to diagnose
• provide prognostic info
• provide initial baseline assessment of functional status
• progress or outcome of therapy

Question 17

What is a standardized assessment?

Answer 17

a published test that has uniformity in administering and scoring it. Detailed instructions included which allows you to collect quantifiable data. Describes the psychometric properties of the tool.

Question 18

What is standardization?

Answer 18

process of taking an assessment and developing a fixed protocol for administering and scoring and completing studies on the psychometric properties of the tool.

Question 19

What is norm-referenced?

Answer 19

scores are obtained from a representative group of people (norms) which informs of average performance as well as range above and below this.

Question 20

What is important when looking at norm-referenced tests?

Answer 20

must be up to date ad reflect the attributes of the population.

Question 21

What is a criteria?

Answer 21

an established principle/standard based on a known standard.

Question 22

What properties should measurements have?

Answer 22

reliability, validity and responsiveness to change

Question 23

What are the characteristics of a good measurement?

Answer 23

• concept being measures is well defined and measured accurately
• everyone gets the same answer using the same info
• everyone gets a different answer when data changes
• everyone knows what the results mean
• error is reduced as much as possible
• measurement strategy can easily be used

Question 24

What are some aspects of clinical utility?

Answer 24

• cost
• time
• training
• energy and effort
• portability
• acceptability

Question 25

What costs must be considered?

Answer 25

- initial outlay: manual, forms, test materials etc. - recurring costs: forms - equipment not included in the test - cost of specific training or expertise

Question 26

What aspects of time should be considered?

Answer 26

- time to administer: poor stamina/attention span = unreliable results - prepare environment - learn test administration - analyse and interpret test results - report writing

Question 27

What must considered for the test to have a good energy and effort level?

Answer 27

- therapist learning to administer the test - mobility and transfers - demands on the assessor and test-taker

Question 28

What aspects of portability must be considered?

Answer 28

- heaviness/bulkiness - Is it easily transportable - can it be used in small places? - multiple parts to the assessment needing to be transported - purchase equipment to carry it?

Question 29

What must be considered in order for the test to be acceptable?

Answer 29

- match of assessment with philosophy and frameworks used in practice context - will the client find this useful? - will it cause stress and anxiety? - does the assessment look professional?

Question 30

What is responsiveness?

Answer 30

ability to detect change

Question 31

How can responsiveness be quantified?

Answer 31

- indicators of effect size - a modified effect size statistic proposed by Guyatt - receiver operating characteristic (ROC) curves

Question 32

How is effect size measures?

Answer 32

cohen's-d

Question 33

What is a small effect size?

Answer 33

0.2 or less

Question 34

What is a moderate effect size?

Answer 34

0.8 or more

Question 35

What scores of effect size indicate responsiveness to clinical change.

Answer 35

The higher the score, the more responsive it is

Question 36

How is effect size calculated?

Answer 36

the differences in means between measurement points (baseline and follow up) divided by measure of variance (standard deviation of baseline)

Question 37

What is Guyatt's responsiveness coefficient (RC)

Answer 37

the smallest difference in a measure that signifies a clinically significant change

Question 38

What is the ROC curve?

Answer 38

reciever operating characteristic curve. True positive rate (sensitivity) is plotted in function of the false positive rate (100-specificity) for different cut-off points

Question 39

What does each point on the ROC curve indicate?

Answer 39

sensitivity/specificity pair corresponding to a particular decision threshold

Question 40

What is specificity?

Answer 40

Probability that a test result will be negative when disease or change isn't present

Question 41

What does the ROC curve look like for a test with perfect discrimination?

Answer 41

curve passes through upper left corner (100% in both axes)