Standardized Tests Flashcards
purpose of developmental testing
Identifying risk of developmental delay
Determine eligibility for services (early intervention or school-based)
Documenting changes
Efficacy of interventions
Research
how to select a test
acceptability
simplicity
cost
appropriateness
reliability
validity
purpose
characteristics of child
content areas assessing
setting it will take place
external constraints
purpose of testing
discriminative/diagnostic
prognostic/screening
evaluative
mean chronological age represented by a test score
easy for parents to understand, but be cautious with delivering this information
age equivalent score
Scores are interpreted on the basis of absolute criteria - how many items were answered correctly
Not how the rest of a “normal group” performed
Measures the child’s mastery of a set of objectives
Less interested in skills being achieved by a particular age, more concerned about the elements that contribute to function
Maximize the validity of the content
Developmental quotient
criterion referenced test
ratio between child’s score (developmental age) and their chronological age
developmental quotient
Uses normative values to interpret test scores
Compared the child and the norm/avg of a group (previously collected scores on a defined population
Items capture the full range of performance on a particular set of knowledge, skills, or abilities
What is the child’s repertoire or behaviors compared to the avg. behavior of a large sample
norm referenced
The number of children of the same age who would be expected to score lower than the child tested
percentile score
The total items passed or correct on a test
May require establishing a basal or ceiling level of performance
raw score
consistency/repeatability between measurements
reliability
percentage of agreement between 2 independent observers
Do 2 testers get the same scores
inter-rater reliability
relationship of a person’s score on the first administration is the same as the second if repeated under identical conditions
test-retest reliability
Indicates if a test measured what it says it is supposed to measure
validity
examines the theory/hypothetical constructs underlying the test
construct validity
appropriateness of the test or how well the content samples the subject matter or behaviors which conclusions are drawn on
content validity
relates the performance on the test to performance on another well-known and accepted test that measures the same knowledge or behavior
concurrent validity
the child’s performance on the test predicts some actual behavior
predictive validity
Measure of reliability
Indicates precision of an individual test score
Estimates the margin of error associated with a test score
Related to the probability of observing a score at a given time
Used to develop confidence intervals (accuracy of a test)
Published in the user manual for each test
standard error of mean
Deviations or variations from the mean
Expressed in units of standard deviation (SD)
standard scores
used to determine if the change in scores from one administration to another is different enough to be “true change”
typically reported according to the confidence interval (90 or 95), also taking into consideration of the standard error of the mean
minimal detectable change (MCD)
smallest amount of change that is meaningful from the patient/therapist
Anchor based: determines the average change of those who have been defined as having improved
Distribution bases: takes the effect size into consideration
Measures may have more than one change
minimal clinically important difference (MCID)
measures with high levels of sensitivity indicate that a negative test result will rule out the diagnosis
sensitivity
measures with high levels of specificity (true positive rate) indicate that a positive test result will run in the condition/diagnosis
specificity
proportion of true positive among all those with positive scores
positive predictive value
