Assessment and Evaluation Flashcards

A research assistant gives the same depression questionnaire to participants two weeks apart and correlates the scores. What property is being assessed?

Answer: Test-retest reliability

Correlating scores from the same instrument given at two time points estimates test-retest reliability.

Cronbach's alpha is most commonly used to estimate which characteristic of a multi-item scale?

Answer: Internal consistency reliability

Cronbach's alpha measures how closely related a set of items are as a group, which is internal consistency.

Two research assistants independently score the same videotaped patient interviews. Which statistic best quantifies their agreement on a categorical rating while correcting for chance?

Answer: Cohen's kappa

Cohen's kappa measures agreement between two raters on categorical data, adjusting for agreement expected by chance.

A new 5-minute memory screening test is compared against a full neuropsychological battery given at the same visit. What type of validity is being established?

Answer: Concurrent validity

Comparing a new measure against a gold standard administered at the same time establishes concurrent criterion validity.

An evaluation conducted during a program's rollout to identify problems and improve delivery is best described as what?

Answer: Formative evaluation

Formative evaluation is done during development or implementation to guide improvements.

A scale consistently reads 2 kg heavier than the true weight for every participant. How would this measurement be described?

Answer: Reliable but not valid

Consistent readings show reliability, but a systematic offset from the true value means the measurement is not valid.

Expert reviewers judge whether a survey's items fully cover all aspects of medication adherence. Which type of validity does this address?

Answer: Content validity

Content validity reflects how well an instrument covers the full domain of the construct, usually judged by experts.