Of Test and Testing Flashcards
a relatively enduring characteristic that distinguishes one individual from another.
Traits
similar to traits but less enduring
States
encompass a wide range of characteristics, including intelligence, cognitive style, adjustment, interests, attitudes, and personality
Psychological traits
Assumption 1
Psychological Traits and States Exist
The meaning of a test score depends on how the trait was defined and what behaviors are considered indicative of it.
Defining and Measuring Traits
The relative importance of different items is determined by technical considerations, the definition of the construct, and societal values.
Item Selection and Weighting
Test scores are derived from examinee responses and are presumed to represent the strength of the targeted ability or trait.
Scoring and Interpretation
Assumption 2
Psychological Traits and States Can Be Quantified and Measured
obtained sample of behavior is used to make inferences about how the individual might behave in real-world situations.
Predicting Future Behavior
psychological tests may be used to understand behavior that has already occurred, such as a criminal defendant’s state of mind at the time of a crime.
Postdicting Past Behavior
tests provide only a sample of behavior and cannot fully reconstruct someone’s state of mind.
Limitations of Testing
Assumption 3
Test-Related Behavior Predicts Non-Test-Related Behavior
Competent test users understand how a test was developed, its intended use, and its limitations.
Understanding Test Development
Test users acknowledge the limitations of the tests they use and understand how those limitations might be compensated for by data from other sources.
Recognizing Limitations
Ethical codes of assessment professionals emphasize the importance of test users knowing the tests they use and being aware of their limitations
Ethical Considerations
Assumption 4
Tests and Other Measurement Techniques Have Strengths and Weaknesses
refers to factors other than what a test attempts to measure that can influence performance
Error
Test scores are always subject to questions about the degree to which error is present.
Error in Measurement
include the assessee’s physical or emotional state, the assessor’s professionalism, and the quality of the measuring instrument itself.
Sources of Error Variance
assumes that each testtaker has a true score that would be obtained if not for the random action of measurement error.
Classical True Score Theory
Assumption 5
Various Sources of Error Are Part of the Assessment Process
Test developers strive to create instruments that are fair when used in accordance with guidelines.
Fairness in Testing
Some fairness-related issues are more political than psychometric, such as debates surrounding affirmative action programs. The question often becomes “What do we as a society wish to accomplish by using this test?”
Political Considerations
It’s important to remember that tests are tools and, like any tool, can be used properly or improperly.
Tests as Tools
Assumption 6
Testing and Assessment Can Be Conducted in a Fair and Unbiased Manner
Tests are essential for making informed decisions in various areas, such as education, employment, and healthcare.
Importance of Testing
Tests help to identify individuals’ strengths and weaknesses, diagnose educational and psychological difficulties, and screen recruits for key variables.
Benefits of Testing
Assumption 7
Testing and Assessment Benefit Society
refers to the consistency of a measuring tool
Reliability
refers to the accuracy of a test
Validity
A __________ measures what it claims to measure
Valid Test
evaluates an individual’s score by comparing it to the scores of a group of testtakers. This helps determine an individual’s standing relative to others.
Norm-Referenced Testing
refers to the number of items answered correctly, multiplied by 100 and divided by the total number of items. It’s not the same as a percentile.
Percentage Correct
rank test scores within a distribution. The xth percentile represents the score at or below which x% of scores fall.
Percentile Ranking
the entire group of interest
Population
a portion of the population that is representative of the whole
Sample
has clearly specified procedures for administration and scoring, typically including normative data. This ensures that all testtakers take the test under the same conditions.
Standardized Test
makes test scores more comparable across different testtakers. It helps ensure that differences in scores reflect actual differences in ability, not variations in testing conditions.
Standardization
assesses an individual’s performance against a predetermined standard or criterion.
Criterion-referenced evaluation
