assessment principles Flashcards
define discriminative measurements
attempts to differentiate between two or more groups of people
define predictive measurements
attempts to classify people into a set of predefined measurement categories for purpose of estimating outcome
define evaluative measurement
pertains to measurement of change in an individual or group over time
define descriptive measurement
pertains to efforts to obtain a ‘clinical picture’ or baseline of person’s skills
what are the 4 types of assessment
non standardised
standardised
criterion referenced
norm referenced
what does measurement enable therapists to do
- quantify attributes of individuals
- make comparisons
- document on performance change
define evaluation
The process of determining the worth of
something in relation to established benchmarks
using assessment information.
define re-evaluation
process of critical analysis of client response
to intervention
define screening
A quick review of the client’s situation to determine if an occupational therapy evaluation is warranted
define testing
a systematic procedure for observing a person’s behaviour & describing it with the aid of a numerical scale or a category-system
define evidence based practice
The integration of best research evidence available, clinical experience and patient values
define non standardised assessments
Do not follow a standard approach or protocol
May contain data collected from interviews, questionnaires and observation of performance
define standardised assessments
- Are developed using prescribed procedures
- Are administered and scored in a consistent manner under the same conditions and test directions
define descriptive assessments
to describe individuals within groups and to characterise differences
define evaluative assessments
use criteria or items to measure an individuals trait over time
define predictive assessments
use criteria to classify individuals to predict trait against criteria
define criterion referenced assessment
client performance is assessed against a set of predetermined standards
define norm referenced assessment
client performance is assessed relative to the other students
pros of criterion referenced assessments
- sets minimum performance expectations
- demonstrates what clients can and can not do
cons of criterion referenced assessments
- hard to know where to set boundary conditions
- lack of comparison data
define norm referenced assessments
Based upon the assumption of a standard normal (Gaussian) distribution with n > 30.
pros of norm referenced assessments
- ensures a spread
- shows client performance relevant to group
cons of norm referenced assessments
- in a strong group, some will be ensured an f
- above average performance is not necessarily good
define reliability
The reproducibility of test results on more than one occasion by the same researcher using a measure.
range from 0 - 1
define random error
errors that can not be predicted
define systematic error
errors that have predictable fluctuations
list the types of reliability
Intra-rater reliability Inter-rater reliability Test-retest reliability / temporal stability Alternate form reliability Split half reliability Internal consistency
intra rater reliability
The stability of data collected by one person more than 2 times
inter rater reliability
Detecting variability between 2 eaters who measure the same client
test retest reliability
The reliability/stability of measurements when given to the same people over time
alternate form reliability
the degree of correlation between two different, but equivalent forms from the same test completed by the same group of people
split half reliability
the degree of correlation between one half the items of a test and the other half of the items of a test (e.g., odd numbered items correlated with the even numbered items)
internal consistency
the the degree of agreement between the items in a test that measures a construct
cronbachs coefficient alpha
used to assess internal consistency; estimate the reliability of scales or commonality of one item in a test with other items in a test; ranges from 0.10-0.99
kappa (k)
used in assessments yielding multiple nominal placements since it corrects for chance
weighted k
used to determine the reliability of a test when rating on an ordinal scale
validity
the extent to which a test measures what it purports to measure
construct validity
Establishes whether assessment measures a construct and its theoretical components
what are the 3 parts of construct validity
- describe the constructs that amount for test performance
- compose hypotheses that explains relationship
- test hypotheses
list the 4 subtypes of construct validity
- convergent
- divergent
- discriminant
- factor analysis
covergent validity
Level of agreement between 2 tests that are being used to measure the same construct
divergent validity
Distinguishing the construct from confounding factors
discriminant validity
The level of disagreement when two tests measure a trait
factor analysis validity
statistical procedure used to determine whether test items group together to measure a discreet construct or variable
content validity
The extent to which a measurement reflects a specific domain
criterion validity
Implies outcome can be used a substitute for ‘gold’ standard criterion test
what are the 2 subtypes of criterion validity
- concurrent/congruent validity (degree to which results agree with others)
- predictive validity (extent to which measure can forecast)
face validity
A test appears to measure what its author intended it to measure
ecological validity
The outcome of an assessment can hold up in the real-world circumstances
what are the 2 types of experimental validity
- internal
- external
sensitivity
Ability of a test to detect genuine changes in a client’s clinical condition or ability
specify
A test’s ability to obtain a negative result when the condition is really absent (a true negative)
responsiveness
providing evidence of the ability of a measure to assess and quantify clinically important change
nominal measurement
only have two response options to items; for example male/female; yes/no wet/dry; happy/sad
ordinal measurement
data has some order, with one score being better/worse than another.
interval scales
the differences between any two scores ratings are identical (such as weight, temperature and distance); statistics can be used correctly.
