PSYC 549: Psychometrics Flashcards
Achievement Test
A test which is designed to measure previous learning. In comparison to an aptitude test, an achievement test assesses knowledge and skills that one already possesses. These tests rely heavily on content validation procedures.
Aptitude Test
A test designed to measure an individual’s potential for learning a specific skill. In comparison to an achievement test, an aptitude test assesses an individual’s ability to potentially acquire certain knowledge and skills. These tests rely heavily on predictive criterion validation procedures.
Construct
A characteristic that varies between individuals but is not directly observable. Constructs are used to explain phenomena in human behavior. Because constructs do not have a concrete existence in the physical world, they cannot be objectively defined and measured.
Criterion-Referenced Scoring/Tests
Tests in which the test-taker is asked to demonstrate a specific skill or ability. Performance is measured against a fixed set of predetermined criteria or standards. These results are not compared to norms.
Criterion-Related Validity
A type of validity in which a test or measure is assessed according to the extent to which is corresponds with a particular criterion or standard. The two types are concurrent, which is how well the test correlates currently with another measure/criterion, and predictive, which is how well the test predicts future performance in relation to a criterion or measure.
Cross-Validation
The process of evaluating a test for a sample other than the one used in the original study. Cross-validation allows the researcher to determine if a test is valid by using a different sample than the initial test sample to see if similar results are found in the second sample.
Norm-Referenced Scoring/Tests
These are tests in which test-taker results are compared to norms. Norms are not standards but are what a typical result should be based on a sample of other results. Tests should be normed on a sample that is reflective of the population that’s tested.
Objective Tests
A type of assessment instrument consisting of a set of items or questions that have specific correct answers, such that no interpretation, judgment, or personal impressions are involved in scoring.
Projective Tests
Tests in which the test-taker is asked to provide a spontaneous response to ambiguous stimuli rather than choosing an answer from provided response options. The theory behind this is that the test taker will project their unconscious conflicts onto the stimuli.
Reliability (types of)
Reliability is the extent to which a test or measure yields consistent results or is free of measurement error. Test-Retest reliability provides a reliability coefficient for the correlation between one person’s scores on one measure taken two separate times. Parallel-Forms reliability develops two separate but equal forms of a test and the scores between them are correlated. Split-Half reliability divides a test’s items into two halves and correlates them. Interrater reliability refers to the correlation between two independent raters scores.
Standard Deviation
This measures the variability within a set of scores or values, indicating how narrowly or broadly they deviate from the mean. The larger the SD, the further a score is from the mean. It is calculated by taking the square root of the sample’s variance.
Standard Scores
Scores which have been standardized to have a fixed mean and standard deviation and to which raw scores can be converted to in order to facilitate comparison and evaluation. Conversion to standard scores allows for more objective comparison across samples/studies and accurate interpretation of said scores.
Test Bias
The tendency of scores on a test to systematically over or underestimate the true performance of individuals to whom that test is administered, particularly because they are members of specific groups (i.e. sex, gender identity, race, etc.). Tests are considered biased if a test design systematically disadvantages certain groups of people over others.
Validity (types of)
Validity measures how well a particular measure fulfills the function for which its being used. Face validity refers to whether a measure appears valid. Content validity refers to how well a measure encompasses the full domain of what it’s trying to measure. Criterion-related validity refers to the extent to which a test corresponds with a particular criterion against which it’s compared – either concurrently or predictively.
Variance
The measure of the spread of scores within a sample or population. A small variance indicates highly similar scores, all close to the sample mean, and a large variance indicates scores spread over a larger range.