Research & Evidence-Based Practice Flashcards
7 cards from real ATP practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Research & Evidence-Based Practice flashcards as text
Which statistical concept describes the degree to which a test score reflects true ability rather than measurement error?
Answer: Reliability coefficient
The reliability coefficient quantifies the proportion of observed score variance attributable to true score variance, directly reflecting measurement precision.
In Item Response Theory, the 'b' parameter of a dichotomous item represents:
Answer: Item difficulty (location)
The b parameter in IRT is the difficulty or location parameter, indicating the ability level at which a test-taker has a 50% probability of answering correctly (in the 1PL/2PL models).
A researcher finds that exam scores predict job performance equally well for male and female applicants. This demonstrates:
Answer: Predictive bias absence
When regression lines for predicting a criterion from test scores do not differ significantly across subgroups, the test shows no predictive bias.
Which research design provides the strongest evidence for causal inference in educational testing?
Answer: Randomized controlled trial
Randomized controlled trials allow causal conclusions because random assignment controls for confounding variables.
The Standards for Educational and Psychological Testing defines 'consequential validity' as primarily concerned with:
Answer: Intended and unintended social consequences of test use
Consequential validity examines the value implications and social consequences—both intended and unintended—that result from test score interpretations and uses.
Differential item functioning (DIF) analysis is primarily used to detect:
Answer: Items that perform differently across groups after matching on ability
DIF identifies items where examinees of equal ability from different groups have systematically different probabilities of a correct response, suggesting potential bias.
Which approach to standard setting requires panelists to estimate the probability that a minimally competent candidate would answer each item correctly?
Answer: Angoff method
The Angoff method asks panelists to estimate the proportion of minimally competent examinees expected to answer each item correctly, then averages these estimates.