ABCN Deck 7 Flashcards
The random errors around the true score have a _____ ______ and have a mean over infinite trials of 0. The standard deviation of the random errors around the true score is termed the ___ ___ __ ___.
normal distribution, standard error of measurement.
What calculation provides an estimate of the reliability of a measure?
variance of true scores/variance of the observed scores
A “tighter” distribution of variability provides (high or low) reliability.
high
In a negative skew, scores tend to cluster at the ____ end of the distribution.
higher
In a positive skew, scores tend to cluster at the ____ end of the distribution.
lower
_______ is focused on item-level characteristics rather than on the test-level characteristics that classical test theory typically involves.
Item response theory
The 3 elements of Baye’s theorem are ___. They produce the ____.
1) prior probability distribution 2) likelihood function 3) available new data; posterior probability.
Bayesian methods are used to assess for ____.
noncredible performance.
Many measures in ___ and ___ have a positive skew.
motor performance, reaction time
the average of the squared difference of each observation in a distribution from the mean.
Variance
the standard deviation is the ___ of the ____.
square root, variance.
Reliability tells ____.
to what degree individual differences can be attributed to true differences in the attribute being tested, versus chance errors.
Reliability ratio?
true variance:total variance
Reliability that looks at the stability of scores on repeated administrations of an instrument to the same person.
test-retest
reliability that captures both the stability of the test over time and the consistency of responses to different samples of items tapping the same knowledge or performance.
alternate forms
reliability that involves splitting the test in many different ways using only a single administration to evaluate internal consistency.
split-half reliability
reliability or consistency that involves estimating 2 sources of error: content sampling & heterogeneity of the domain of knowledge/behavior.
inter-item
the ___ formula results in the reliability coefficient for internal consistency of a dichotomously scored test and represents the mean of all half-reliabilities resulting from splitting the test.
Kuder-Richardson (KR20)
___ is the accuracy with which meaningful and relevant measurements can be made with a test.
Validity
___ validity is the degree to which the test covers a representative sample of the knowledge or behavior being studied.
Content
___ validity is the degree to which a test measures what it was intended to measure by comparing it to previously validated tests.
concurrent
___ validity is the degree to which a test, measurement or protocol successfully measures a psychological theoretical construct or trait.
construct
threats to internal validity are related to ___
conclusions drawn from the use of the test.
threats to external validity related to ___
limits on generalizability of test results.
Threats to internal validity include (3)
history (partially addressed with appropriate norms), testing interval, order of test administration.
Threats to external validity (3)
regression to the mean, multiple comparisons (some low scores may be due to chance), situational variables (medication, mood, task engagement)
____ validity is demonstrated when two or more approaches to measurement of some trait are positively correlated.
convergent
The ____ is a composition of correlation coefficients of 2 or more traits and two or more methods.
multitrait multimethod matrix
the ability of a test to correctly identify patients with a disease
sensitivity
the ability of a test to correctly identify people without the disease.
Specificity
The likelihood ratio of a positive test compares ___ to ___.
true positives to false positives
The likelihood ratio of a negative test compares ___ to ___.
false negatives to true negatives
Positive likelihood ratio values great than 1 suggest that a positive test result is___.
indicative of the presence of a condition.
A negative likelihood ratio between 0 and 1 indicates ___.
few false negatives as compared to true negatives.
the probability that a specific individual has a specific condition, given that a test finding was positive, can be calculated using ___.
Baye’s theorem.
___ validity is the extent to which a test improves the post-test probability with respect to the pre-test probability. It is the positive predictive value of the test minus the baserate of the condition.
incremental
If our goal is to identify all persons showing any degree of impairment, we would want to choose a cut-off score that maximizes ___.
sensitivity
If you want to minimize false positive errors, such as in effort testing, you would want to choose a cut-off score that maximizes ___.
specificity
As the reliability of a test decreases, the estimated true score is brought ___ to the mean and the confidence band ___.
closer, widens.
As the reliability of the test increases, the __ and ___ and therefore the confidence intervals both decrease.
SEE and SEM
____ ___ informs us of the statistical significance of the difference between two scores.
Reliable difference
The __ __ __ establishes the minimum magnitude of change required for psychometric certainty that two scores actually differ.
reliable change index
__ refers to whether an intervention produces the expected result under ideal circumstances.
efficacy
measures the benefit of an intervention under real word conditions.
effectiveness
The ____ reflects the probability that the measured change is reliable; an observed difference between 2 scores from the same examinee on the same test cannot be attributed to measurement error.
reliable change index
Calculation of the RCI is___. An RCI falling more than a z-score of ___ reflects a significant difference in test scores.
a z-score for the difference between the individual tests based on the normal probability distribution. 1.96.
Reliability change indexes should be statistically adjusted for practice effects using the _____.
mean change score for the reference group.
___ is the process of of utilizing a score profile to determine whether an individual belongs to one group or another.
discriminant analysis
a _____ is a capacity shared by all healthy members of the species.
species comparison standard
___ require the person’s level of ability before the onset of known or suspected neurological dysfunction be determined and used to compare current scores against.
individual comparison standards
A negative test score on a test with high sensitivity ____ a diagnosis.
rules out a diagnosis
a positive result on a test with high specificity ____ a diagnosis.
rules in a diagnosis.
the ____ is the estimate of how much a positive test result will change the odds of having the disease.
likelihood ratio positive
the ___ is the change in odds of not having a disease when a test results is negative.
likelihood ratio negative
LR > 1 indicates a strong likelihood that___.
a disease is present
LR < 0.01 indicates that a test result is associated with ____.
the absence of a disease.
the ____ is the probability of the patient having the disease with a positive diagnostic test result.
post-test probability
As a rule of thumb, a good diagnostic test will increase the post-test odds to above ___ in populations in which the prevalence rate is low.
0.50
When a battery includes 36 variables, based on the full NAB, ___% of the healthy normative sample obtained two or more scores below the 5th percentile. __% had five or more scores below the 5th percentile.
48.5, 15
Among individuals with FSIQ 90 to 109, __% to ___% of subjects had at least one score below the 5th percentile.
22-38