Study sets, textbooks, questions
Upgrade to remove ads
Terms in this set (47)
"Extend to which measurement is consistent and free from error"
Measured score =
True score + or - error
Types of measurement error
Always overestimate/underestimate - for example - starting on the wrong point on a tape measure
Due to chance, from measurement to measurement are unpredictable
Regression to the mean (meaning_
If you take enough measurements random measurement erros are normally distributed so they will equal out over time and sum to 0
Sources of measurement error
Rater error - in perception or reporting measurement
Instrument - Faulty function of mechanical instrument
Variability of characteristic being observed - natural variability such as blood glucose changes over time
Ratio of total variability between scores to variability within scores
true score variability/true score variability + error variability
If it is 0 no reliability
If 1 = perfect reliability
Indicates how much of a measured value, expressed in the original units, is likely due to error
If true variance makes up a large proportion of total variance reliability is
General reliability standards
ICC> .9 for clinical measurements
>.75 = good reliability
less than that is poor to moderate
Between 1 person
a method for determining the reliability of a test by comparing a test taker's scores on the same test taken on separate occasions
If you do pain scale and don't retest for a long time, unlikely that it is reliable because things could have changed
Alternate forms reliability
To see if 2 versions of same instrument are equivalent - uses same unit of measurement
Most often measured using Cronbach's alpha
Estimates how well items that reflect the same construct yield similar results such as in a questionnaire
Change scores are used to do what?
Demonstrate effectiveness of an intervention
Track course of disorder over time
Provides context for clinical decision making
Floor effect or ceiling effect
Ceiling effect - can't detect change if test is too low level
Floor effect -
the ability of an instrument to detect small but meaningful change over time
Minimal detectable change (MDC)
Ability of instrument to detect change beyond measurement error
Is this change real or just measurement error?
MCID (Minimal clinically important difference)
Ability of an instrument to detect minimally important change
Context dependent - 10degrees ROM change not much with hip flex, but 1/3 of norm in hip ext
Is this change meaningful to my patient?
Validity and reliability: which can you not have without the other?
You can't be valid without being reliable
extent to which an instrument measures what it is intended to measure
A test can be _____ but not ______
For example, a scale off by two pounds too high
Reliable but not valid
An instrument appears to test what is intended to test - but only after the test is developed. This judgement is made by those who use the instrument
Ex: ROM, strength, balance
•Do measurements adequately represent concept (and not concepts that are unrelated)?
•Typically used in questionnaire development
There are no stats to measure this, is only determined by expert review
Three requirements for content validity
1.Items must represent the full scope of the construct being studied
2.Number of items are proportionate to the importance of that component
•Ex: Topics on DPT 653 Exam 1 are proportionate to the amount of time spent on the topic in class
3.No irrelevant items
Can the outcomes of the instrument be substituted for an established gold standard?
measured by correlation coefficients and sensitivity/specificity analyses
What is the highest and most objective form of validity?
What is the lowest form of validity
Two types of criterion related validity
•Measurements between tests taken at approximately same time
•DASH vs. Quick DASH
•Establishes that the outcome of the target test can be used to predict a future criterion score or outcome
•Do GREs predict 1st year success in grad school?
•Reflects the ability of an instrument to measure the theoretical dimensions of a construct
How well a tool measures an abstract concept such as depression, disability etc
Methods of construct validation
Convergence and divergence
Factor analysis - not tested
Known groups method
•Do test results differ between 2 groups (known to be different)?
•E.g. Is new pain score different between acutely injured and people 2 months after injury?
•Do test results correlate with other similar tests?
•E.g. Pain and disability
•Low correlation with different tests?
•E.g. New pain score and height
•Extent that sample systematically misrepresents population
•Can be conscious (e.g. email only dry needling practitioners)
•Can be unconscious (e.g. people only respond if they have a strong opinion)
•Extent that sample randomly misrepresents population
•"The difference between sample averages (called statistics) and population averages (called parameters)" (p. 183)
•Estimations of sampling error are key to all inferential statistics
•Random selection (or other known probability)
•Sample is "considered" representative of population (although it may not be)
•Can estimate sampling error
•Generalization is difficult
•Cannot estimate sampling error (therefore officially should not use inferential statistics)
Types of probability sampling
•Simple Random Sampling
--Ex: select every 10th person
•Stratified Random Sampling
--Specify number from each category
Types of nonprobability sampling
--Subjects chosen on basis of availability
-Potential bias of self-selection?
-Like stratified, but not random
-Subjects hand-picked by specific criteria (e.g. case series)
Recommended textbook explanations
Psychology: Principles in Practice
Spencer A. Rathus
Myers' Psychology for AP
David G Myers
Myers' Psychology for AP
David G Myers
A Concise Introduction To Logic (Mindtap Course List)
Lori Watson, Patrick J. Hurley
Sets with similar terms
Reliability and validity
USA Scientific Inquiry Exam 2
Research methods quiz 2
Other sets by this creator
Pro D 2
Anatomy Hand/Exam 1
Human Imaging wk3
What do we call the kind of learning in which behavior is strengthened if followed by a reinforcer? a. Operant conditioning. b. Respondent behavior. c. Classical conditioning. d. Shaping. e. Punishment.
Brain scans and EEG recordings indicate that positive emotions are associated with high levels of activity in which brain section? a. Right temporal lobe. b. Cerebellum. c. Left frontal lobe. d. Left temporal lobe. e. Right parietal lobe.
An individual with an exceptional memory is identified. For any given date, she is capable of recalling major events, the weather, and what she did that day. What research method is being used if a psychologist conducts an in-depth investigation of this individual using questionnaires, brain scans, and memory tests? a. Naturalistic observation. b. Survey. c. Interview. d. Case study. e. Theory method.
Emily's family has been traumatized by a recent hurricane in Puerto Rico. Described in terms of the general adaptation syndrome (GAS), which phase would they have experienced first? a. Resistance b. Appraisal. c. Exhaustion. d. Alarm. e. Challenge.
Other Quizlet sets
HIST 1302 Exam 1
practice of real estate
PTA128 Chapter 9 - Maintaining Capacity for Mobili…
Aquatic Science U3