3. Psychological Testing
Definition of a Test
Based on J.M. Cattell (), “Mental test”
Standardized procedure to sample behavior and describe it with scores/categories.
Most tests include norms/standards for predicting other behaviors.
Defining features: Standardized procedure, Behavior sample, Scores/categories, Norms/standards, Prediction of non-test behavior.
Standardized procedure
Uniform administration procedures across examiners/settings.
Depends on examiner competence and manual directions.
Clear directions are crucial for standardization.
Behavior Sample
Limited sample of behavior provides inferences about a broader domain.
Test items don't need to resemble predicted behavior.
Key: permits prediction of other behaviors.
Scores or categories
Furnishes scores or categorizes individuals.
Cautions: traits are vague, scores have measurement error.
Consumers must avoid reifying the characteristic being measured.
Norms or standards
Norms: summary of test results from a large, representative “standardization sample.”
Allows generalization beyond the test, predicting real-world functioning.
(Exception: criterion-referenced tests).
Prediction of non-test behavior
Purpose: predict behaviors beyond those directly sampled.
Example: inkblot response predicting suspiciousness.
Prediction ability validated by research after test release.
Distinction in Testing
Norm-referenced tests: performance compared to a standardization sample.
Criterion-referenced tests: performance assessed against specific educational objectives.
Objective measurement
Objectivity in administration, scoring, and interpretation, independent of examiner's subjective judgment.
Item difficulty based on objective, empirical procedures.
Differences of test content
Content: subject matter varies by purpose.
Format: type, structure, item count.
Selected-response (forced-choice): true-false, multiple-choice.
Constructed-response: fill-in-the-blank, essays, performance tasks.
Administration: procedures vary.
Scoring/Interpretation: hand, computer, publisher, self-scored; transforming data into meaningful info.
Cost.
Types of Tests based on administration procedures
Group tests: large groups, often pencil-and-paper.
Individual tests: one-on-one, allows gauging motivation.
Other types: self-administered, computer, video/audio, nonverbal.
Types of tests
Many types measure different aspects; categorization simplifies.
Categories of tests
8 categories:
Intelligence: global abilities (verbal, perceptual, reasoning); predicts scholastic/occupational potential.
Aptitude: capability for specific tasks/skills; narrow ability.
Achievement: learned knowledge/accomplishment.
Creativity: novel thinking, unusual solutions for vague problems.
Personality: traits, qualities, behavior (checklists, inventories, projective).
Interest inventories: preferences for activities/topics; aids occupational choice.
Behavioral procedure: describes/counts behavior frequency; identifies antecedents/consequences.
Neuropsychological: cognitive, sensory, perceptual, motor performance; assesses brain damage effects.
Uses of Testing
5 uses:
Classification,
Diagnosis/treatment,
Self-knowledge,
Program evaluation,
Research.
Uses of testing: classification
Assigning people to categories for:
Placement: sorting into programs (needs/skills).
Screening: quick ID of special characteristics.
Certification/Selection: pass/fail for privileges/proficiency.
Uses of testing: diagnosis
Two tasks:
Determine nature/source of abnormal behavior,
classify within diagnostic system.
Should include strengths, weaknesses, etiology, and treatment plan.
Factors influencing the soundness of testing
Influenced by many factors:
administration,
tester characteristics,
context,
examinee motivation/experience,
scoring method.
These can alter results.
Standardized procedures in test administration
Reliable interpretation requires standardized conditions per manual.
APA standards guide valid testing.
Follow procedures (instructions, timing, materials); exceptions only with professional judgment (clinical).
Desirable procedures of test administration
Individual testing:
Examiners must know materials/directions.
Details are vital for valid results.
Sensitivity to disability: impairments distort results; specialized tests needed.
Desirable procedures of group testing
Errors: incorrect timing (too little/too much), unclear directions, uncontrolled noise, not explaining guessing.
Deviations from manual unacceptable in group testing.
Influence of the examiner
Rapport is crucial: comfortable atmosphere motivates cooperation.
Poor rapport leads to anxiety, noncooperation, hostility, distorted results.
Examiners vary; avoid cold/overly solicitous styles.
Examiner sex, experience and race
Studies are contradictory/inconclusive.
Mostly show little effect of examiner sex, experience, race on scores.
Still, specific characteristics might have significant effects.
Background and motivation of the examinee
Results distorted by: anxiety, malingering, coaching, cultural background.
Test Anxiety
Responses (psychological, behavioral) linked to concern about test failure.
Responsibilities of test publisher
Publisher: Guard against premature release.
Standards: APA sets standards for technical manuals/user guides.
Manuals: Must detail reliability, validity, normative sample, technical stats.
Marketing: Responsible advertising; accurate info in manuals.
Competencies of test purchasers
Access limited by test complexity.
APA defines 3 levels of expertise (, , ).
Level A
Simple paper-and-pencil tests.
Minimal training needed; usable by non-psychologists (e.g., executives, educators).
Examples: vocational proficiency, group achievement tests.
Level B
Requires test construction, statistics, psychology knowledge.
For those with advanced testing course or psychologist supervision.
Examples: aptitude tests, personality inventories (normal populations).
Level C
Substantial testing knowledge, supervised experience essential.
For master’s degree holders in psychology/allied fields.
Examples: projective personality, neuropsychological batteries.
Responsibilities of Test Users
Key practices:
Client's best interests.
Confidentiality and duty to warn.
Test user expertise.
Informed consent.
Avoid obsolete tests; maintain standard of care.
Responsible report writing.
Clear communication of results.
Consider individual differences.
Test Bias and Use with Minority Groups
Bias: systematic measurement error for certain factors/groups.
Criticism: psychological testing often biased towards European American, middle-class society.
Impact of cultural background on test results
Cultural background impacts assessment.
Acknowledge differences, avoid overgeneralization.
Stereotype threat: concern about confirming negative group stereotypes.
Selecting Psychological Test
Key factor: usefulness for referral question.
Others: practitioner's training, experience, preference, literature alignment.