Chapter 6 part 2 Flashcards
The extent to which a test provides an unbiased measurement across different groups.
Fairness –
A type of bias where different groups have the same slope but different intercepts in regression predictions.
Intercept Bias –
A type of bias where different groups have different slopes in regression predictions.
Slope Bias –
A statistical method used to detect bias in test items.
Differential Item Functioning (DIF) –
A method of assessing an individual’s performance or traits using a scale.
Rating –
A scale used to assign scores based on performance or behavior.
Rating Scale –
Inaccuracy in test results due to biases or inconsistencies in scoring.
Rating Error –
A rater error where the evaluator is overly lenient
Leniency Error –
A rater error where the evaluator is overly harsh in scoring.
Severity Error –
A rater error where the evaluator is overly generous in giving high scores.
Generosity Error –
When a rater avoids extreme ratings and gives most ratings near the middle of the scale.
Central Tendency Error –
When a rater’s overall impression of a person influences specific ratings.
Halo Effect –
The proportion of cases in which a test correctly predicts an outcome.
Hit Rate –
The proportion of cases in which a test incorrectly predicts an outcome.
Miss Rate –
A framework outlining the content and structure of a test.
Test Blueprint –
The proportion of individuals in a population who exhibit a particular characteristic.
Base Rate –
A method of scoring individuals by ordering them based on a given characteristic.
Ranking –
A cost-benefit analysis of using a test in decision-making.
Utility Analysis –