Vocabulary Terms

Front 1

Ability

Back 1

A defined domain of cognitive, perceptual, psychomotor, or physical functioning.

Front 2

Bias

Back 2

In a statistical context, a systematic error in a score. In discussing fairness, bias refers to variance due to contamination or deficiency that
differentially affects the scores of different groups of individuals

Front 3

Concurrent validity evidence

Back 3

Demonstration of the relationship between
job performance and other work outcomes, and scores on selection
procedures obtained at approximately the same time.

Front 4

Confidence Interval

Back 4

An interval between two values on a score scale within which, with specified probability, a score or parameter of interest is
expected to lie.

Front 5

Construct

Back 5

A concept or characteristic of individuals inferred from empirical evidence or theory

Front 6

Criterion

Back 6

A measure of work performance or behavior, such as productivity, accident rate, absenteeism, tenure, reject rate, training score, and
supervisory ratings of job relevant behaviors, tasks or activities.

Front 7

Criterion-related validity evidence

Back 7

Demonstration of a statistical relationship between scores on a predictor and scores on a criterion measure.

Front 8

Cross-validation

Back 8

The application of a scoring system or set of weights
empirically derived in one sample to a different sample from the same
population to investigate the stability of relationships based on the original weights

Front 9

Cutoff score

Back 9

A score at or above which applicants are selected for further
consideration in the selection process. The cutoff score may be established on the basis of a number of considerations (e.g., labor market,
organizational constraints, normative information). Cutoff scores are
not necessarily criterion referenced, and different organizations may
establish different cutoff scores on the same selection procedure based
on their needs.

Front 10

Fairness

Back 10

There are multiple perspectives on fairness. There is agreement that
issues of equitable treatment, predictive bias, and scrutiny for possible
bias when subgroup differences are observed are important concerns in
personnel selection; there is not, however, agreement that the term “fairness” can be uniquely defined in terms of any of these issues.

Front 11

Generalized evidence of validity

Back 11

Evidence of validity that generalizes to
setting(s) other than the setting(s) in which the original validation evidence was documented. Generalized evidence of validity is accumulated through such strategies as transportability, synthetic validity/job
component validity, and meta-analysis.

Front 12

Internal consistency reliability

Back 12

An indicator of the reliability of a score
derived from the statistical interrelationships of responses among item
responses or scores on different parts of an assessment.

Front 13

Multiple-hurdle model

Back 13

The implementation of a selection process whereby two or more separate procedures must be passed sequentially

Front 14

Predictive bias

Back 14

The systematic under- or overprediction of criterion performance for people belonging to groups differentiated by characteristics not relevant to criterion performance.

Front 15

Predictive validity evidence

Back 15

Demonstration of the relationship between
selection procedure scores and some future work behavior or work outcomes.

Front 16

Predictor

Back 16

A measure used to predict criterion performance.

Front 17

Reliability

Back 17

The degree to which scores for a group of assessees are consistent over one or more potential sources of error (e.g. time, raters, items,
conditions of measurement, etc.) in the application of a measurement
procedure.

Front 18

Restriction of range or variability

Back 18

Reduction in the observed score variance of a sample, compared to the variance of an entire population, as
a consequence of constraints on the process of sampling.

Front 19

Skill

Back 19

Level of proficiency on a specific task or group of tasks

Front 20

Standardization

Back 20

(a) In test construction, the development of scoring
norms or protocols based on the test performance of a sample of individuals selected to be representative of the candidates who will take
the test for some defined use; (b) in selection procedure administration, the uniform administration and scoring of a selection procedure
in a manner that is the same for all candidates

Front 21

Synthetic validity evidence

Back 21

Generalized evidence of validity based on previous demonstration of the validity of inferences from scores on the selection procedure or battery with respect to one or more domains of work
(job components); also referred to as “job component validity evidence.”

Front 22

Systematic error

Back 22

A consistent score component (often observed indirectly), not related to the intended construct of measurement.

Front 23

Type I and Type II errors

Back 23

Errors in hypothesis testing; Type I error
involves concluding that a significant relationship exists when it does
not; Type II error involves concluding that no significant relationship
exists when it does.

Front 24

Validation

Back 24

The process by which evidence of validity is gathered, analyzed, and summarized. (Note: laypersons often misinterpret the term
as if it implied giving a stamp of approval; the result of the research might be zero validity.)

Front 25

Validity

Back 25

The degree to which accumulated evidence and theory support
specific interpretations of scores from a selection procedure entailed by
the proposed uses of that selection procedure.