[Handouts] JJPangay

Lesson 1 – Introduction to Psychological Measurement

Ancient Roots of Testing

• Chinese Empire
  • Civil-service examinations to rank applicants for government posts.
• Greeks
  • Tests gauged intelligence & physical skills (e.g., Olympic-style trials).
• Medieval/early-modern European universities
  • Written & oral exams for degrees & honors.

Recognition of Individual Differences

• Charles Darwin
  • No two humans identical; some traits more adaptive ⇒ survival & complexity.
• Francis Galton
  • Founded testing movement.
  • Anthropometric lab; rating scales, questionnaires, free association.
  • Created Galton Bar (visual length) & Galton Whistle (audible pitch).
  • Statistical tools for test data; noted sensory deficits in intellectual disability.

Early Experimental Psychologists

• J. F. Herbart – mathematical mind models; pedagogy founder.
• E. H. Weber – sensory thresholds; \text{JND}.
• G. T. Fechner – psychophysics; \text{Weber–Fechner law} linking stimulus & sensation.
• W. Wundt – first psych lab (Leipzig, 1879).
• E. Titchener – brought Structuralism to U.S. (Cornell; his brain on display!).
• G. M. Whipple – seminars on human-ability tests.
• L. L. Thurstone – factor analysis pioneer; law of comparative judgment.

Study of Mental Deficiency & Intelligence

• J. Esquirol – distinguished mental retardation from insanity.
• E. Seguin – educational methods for intellectual disability.
• J. M. Cattell – coined “mental test”.
• A. Binet – father of IQ tests (Binet-Simon scale).
• L. Terman – Stanford revision; coined \text{IQ}=\dfrac{\text{MA}}{\text{CA}}\times100.
• C. Spearman – g & s two-factor theory.
• Thurstone – Primary Mental Abilities.
• D. Wechsler – WAIS, WISC.
• R. Cattell – Fluid gf vs. Crystallized gc.
• J. P. Guilford – 6\times5\times6=180 ability structure.
• Vernon & Carroll – hierarchical g.
• Sternberg – triarchic (academic, practical, creative).
• H. Gardner – multiple intelligences.
• H. Goddard – French Binet-Simon translation (mis-used at Ellis Island).

WW I Contributions

• R. Yerkes – Army Alpha (literate) & Beta (illiterate) mass intelligence tests.
• A. S. Otis – multiple-choice formats.
• R. S. Woodworth – Personal Data Sheet (proto-MMPI) for “shell-shock”.

Personality Testers

• H. Rorschach – Inkblot Test.
• H. Murray & C. Morgan – TAT.
• 1940s – structured inventories rise.
• R. B. Cattell – 16 PF.
• McCrae & Costa – Big 5.

Philippine Pioneers

• V. Enriquez – PUP (“Panukat ng Ugali …”).
• A. Palacio – PKP.
• A. Carlota – PPP.
• G. Del Pilar – Mapa ng Loob.
• A. Lagmay – PTAT.

Psychological Testing & Assessment Basics

Objectives of Psychometrics

• Measure overt/covert behavior.
• Describe & predict traits, states, interests, etc.
• Detect dysfunction for diagnosis & intervention.

Testing vs. Assessment

Core Assumptions

1. Traits & states exist.
2. They are measurable.
3. Test behavior predicts non-test behavior (post-/pre-dict).
4. Tools have strengths & limits.
5. Error is inherent (error variance).
6. Fair, unbiased use is possible.
7. Testing benefits society.

Key Parties

• Authors/Developers, Publishers, Reviewers, Users, Sponsors, Takers, Society.

Referral Settings & Typical Questions

• Psychiatric wards, general medicine, legal (competency, custody, dangerousness), education (LD, placement), clinics.

Nomothetic vs. Idiographic Approaches

• Nomothetic = general laws across persons.
• Idiographic = unique constellation per person.

Inference Metrics

• Base rate, Hit rate (Sensitivity \& Specificity), Misses (False + / -), Cut-scores.

Cross-Cultural Testing

• Parameters: language, content, education, tempo.
• Culture-free ideal impossible; moved to culture-fair with common content.
• Culture loading = degree of cultural content.

Lesson 2 – Legal & Ethical Foundations

Ethics

• Framework of agreed morals guiding professionals.
• Professional ethics: confidentiality, competence, non-maleficence, accountability.

Universal Values

Autonomy, Beneficence, Justice, Fidelity, Honesty, Non-maleficence, etc.

Law vs. Ethics

• Law sets minimum; ethics sets ideal.

Philippine Legal Bases

• \text{RA 10029} (2009) Psychology Act
  • Defines Psychological Interventions, Assessments, Programs.
• PRB of Psychology Res. 11 (2017) – National Code of Ethics.
  • 4 principles: Dignity, Competent Caring, Integrity, Responsibility to Society.
• Res. 12 (2017) – Adoption of international core competencies.

Common Ethical Dilemmas

• Breaking confidentiality, releasing reports, assessing public figures, multiple relationships, gift-giving, labeling, divided loyalties, dehumanization.

Ethical Standards in Assessment

• Publisher responsibilities (quality, restricted sales).
• User duties: best interest, informed consent components, avoid harassment, duty to warn, cultural fairness.
• Tool selection criteria: relevance, validity, familiarity, adaptability.
• Need for batteries (no single test sufficient).
• Administration principles: study manual, maintain rapport, neutral demeanor, control conditions, standard directions, observe guessing policy.
• Rights of Test Takers: courtesy, appropriate tools, results, least stigmatizing label, informed consent.

Lesson 3 – Assessment Techniques & Tools

Traditional Clinical Modality (6-Step)

1. Referral & question
2. Set assessment goals
3. Choose decision standards
4. Collect data
5. Make decisions
6. Communicate results

Tele-Assessment Models

• Technician-Assisted, Hybrid, Modified F2F, Direct-to-Home – differ on location, distancing, tech cost, test restrictions.

Interviews

• Structured, Semi, Unstructured.
• Kanfer & Grimm: focus on behavioral deficits/excesses & contingencies.
• BASIC ID model (Behavior, Affect, Sensation, Imagery, Cognition, Interpersonal, Drugs).
• Uses: diagnosis, therapy targets, risk assessment, contracting.
• Tactics: clarification, probing, confrontation, reflections, summaries; manage eye contact, self-disclosure.
• Preliminaries: physical setup, introductions, purpose, confidentiality limits, process explanation, fees.
• Mental Status components: appearance, speech, affect/mood, thought process/content, memory, etc.
• Widely used structured interviews: SCID, DIS, ADIS, SIDP, etc.

Documents & Portfolios

• Portfolio examples for hiring; case history for developmental/medical background.

Behavioral Assessment

• Direct observation, self-monitoring.
• Recording systems: narrative, interval, event, ratings.
• Analogue studies, situational performance tests, role-play, psychophysiological indices, unobtrusive measures.
• Observer errors: Reactivity (Hawthorne), Drift, Contrast.

Psychological Tests

• Measurement vs. Assessment vs. Evaluation.

Three-Tier User Qualification

1. Level A – basic (achievement tests).
2. Level B – technical knowledge (aptitude, normal personality).
3. Level C – advanced, license (projectives, individual IQ).

Variable-Based Categories

• Ability tests (Intelligence, Achievement, Aptitude) – right/wrong answers, max performance.
• Typical-performance tests (personality, interests, values) – no right/wrong, honesty focus.

Specific Test Types

Intelligence, Aptitude, Achievement, Personality (with theory-guided, factor-analytic, criterion-keyed inventories), Interests, Attitudes, Values, Diagnostic, Power vs. Speed, Creativity, Neuropsychological, Objective vs. Projective, Norm- vs. Criterion-referenced.

Clinical Differences Objective vs. Projective (task definiteness, scoring objectivity, response types, validation emphasis).

Test Uses

• Classification (placement, screening, certification, selection).
• Diagnosis & treatment planning.
• Self-knowledge, program evaluation (diagnostic, formative, summative), research.

Lesson 4 – Psychometric Properties & Statistics Refresher

Scales of Measurement

• Nominal, Ordinal, Interval, Ratio (plus comparative scales: paired, rank-order, constant-sum, Q-sort; non-comparative: graphic, Likert, semantic differential, Stapel).

Descriptive Statistics

• Frequency distributions.
• Central tendency: Mode (nominal), Median (ordinal or skewed interval/ratio), Mean (normal interval/ratio).
• Variability: Range, \text{IQR}, \text{SD}.
• Location: Percentile =\dfrac{#\text{beaten}}{N}\times100, quartiles, deciles.
• Shape: Skew (+ / −), Kurtosis.

Standard Scores (Normal Curve)

• z=\dfrac{X-\bar X}{SD} (Mean 0 SD 1).
• T=10z+50; Stanine =2z+5; Sten =2z+5.5; IQ \mu=100,\,\sigma=15; CEEB \mu=500,\,\sigma=100.

Parametric vs. Non-parametric

• Parametric needs normality, homogenous variances, interval/ratio.
• Non-parametric for ordinal/nominal.

Common Tests

• Correlations: Pearson, Spearman, Kendall, Phi, Lambda.
• Prediction: Biserial, Point-biserial, Logistic, Linear & Multiple regression, Tetrachoric, Ordinal Regression.
• Chi-square (GOF & independence).
• Group comparisons: t-tests, Wilcoxon, Mann-Whitney, ANOVA, Friedman, Kruskal-Wallis.
• Factor Analysis: PCA, EFA, CFA.

Reliability

• Goal: estimate error & improve measurement.
• Sources of error: scorer, time sampling, content sampling, inter-item, combined.
• Coefficients:
  • Test–retest (r_{tt}) – \text{time sampling}.
  • Parallel/Alternate forms – content + time.
  • Split-half (use Spearman-Brown). Internal consistency: KR{20}, KR{21}, Cronbach \alpha, APD.
  • Inter-rater: Cohen \kappa, Kendall W.
• Interpretation: \ge0.90 (excellent clinical); 0.70 minimum research; SEM =SD\sqrt{1-r} ⇒ build 95\% CI \bar X\pm1.96SEM.

Validity

• Face, Content (blueprints, construct under-representation/irrelevant variance), Criterion-related (Concurrent, Predictive, Incremental; requires valid criteria), Construct (homogeneity, development, contrasted groups, convergent/divergent, factor analysis, cross-validation).
• Bias vs. fairness (rating errors: severity, leniency, central tendency, halo, primacy, recency, impression management, acquiescence, faking).

Norms

• Developmental (MA, IQ, grade equiv.), Within-group (percentile, standard scores), Relativity norms (national, local, co-norms, subgroups).

Lesson 5 – Test Utility & Decision Making

Utility Analysis

• Balances Costs vs. Benefits to determine practical value of testing.
• Expectancy data: probability of success for score ranges.
• Taylor–Russell tables incorporate validity, selection ratio = \dfrac{#\text{to hire}}{#\text{applicants}}, and base rate (current success%).

Cut-Score Methods

1. Angoff – SME judges estimate % minimally qualified who answer item correctly ⇒ average.
2. Known-Groups – choose score that best separates criterion groups.
3. IRT-based
   • Item-mapping histogram.
   • Bookmark: SMEs place marker in ordered item booklet.
4. Predictive/Discriminant analysis (optimization given selection needs).

Compensatory Selection Models

• High score in one attribute can offset low in another.

Lesson 6 – Test Development & Standardization

Psychometric Theory Families

• Observed-score: Classical Test Theory (X=T+E), Generalizability.
• Latent-variable: Factor analysis (now CFA/SEM), IRT (1-,2-,3-PL), Rasch (1-PL with fixed discrimination), Mixed-models.

Test Development Approaches

1. Rational/Theoretical – item creation guided by theory/expert logic.
2. Factor-Analytic – retain items loading on factors.
3. Empirical/Criterion-Keyed – keep items that discriminate groups (e.g., MMPI).
4. Projective – ambiguous stimuli to elicit projection.

Standardization Principles

• Control extraneous variables so person factor stands out.
• Uniform test conditions (lighting, noise, motivation), administration (exact instructions), scoring & interpretation rules.
• Standardization sample must represent target population.
• Examiners: meet qualification; prepare materials; orient proctors.

Objectivity in Tasks

• Time-limit: equal time.
• Work-limit: equal workload.
• Manage guessing effects.

Sequential Stages

1. Conceptualization (objectives, constructs, population, blueprint).
2. Construction (write items: one idea each, clear, avoid double negatives, extreme absolutes; choose formats – MCQ, T/F, etc.).
3. Try-out (pilot).
4. Item Analysis (see below).
5. Revision & finalization.

Item Analysis Metrics (Classical)

• Item Difficulty p=\dfrac{Nu+Nl}{N} (optimal ≈ 0.50; retain 0.21–0.80).
• Item Discrimination D=\dfrac{Nu-Nl}{\frac12 N}; accept \ge0.30.
• Item Reliability & Validity indices (item–total correlations).
• Distractor analysis: wrong options should attract more low-scorers.
• Decision grid: unacceptable difficulty or discrimination ⇒ discard or revise.

Psychological Report Writing (Lesson 11)

Purpose & Audience

• Communication bridge from complex data to referral question.
• Must match reader’s background (physician, judge, parent, client).

Qualities of a Good Report

1. Clarity
   • Specific, concise language; short sentences.
   • Logical flow.
2. Meaningfulness
   • Focus on client’s unique functioning; avoid vague generalities.
3. Synthesis
   • Integrate findings into coherent personality map; relate symptoms to underlying system; guide treatment.

Style Guidelines

• Use professional, everyday words with precise meaning.
• Short paragraphs; group similar concepts.
• Include only information that aids understanding/decisions; avoid overload.
• Translate technical jargon where possible; ensure reader comprehension.

Levels of Interpretation

1. Level 1 – Minimal inference; purely descriptive (e.g., aptitude score used for hiring).
2. Level 2 – Descriptive generalizations & hypothetical constructs (e.g., “depressed” based on behaviors).
3. Level 3 – Full theory of person (psychobiography linking development, context, prediction).

Common Interpretive Errors

• Information overload
• Schematization (oversimplifying)
• Insufficient internal/external evidence
• Over-interpretation (wild analysis; fixed symbol meanings)
• Lack of individualization (Barnum statements)
• Poor integration (contradictory traits)
• Over-pathologizing & over-psychologizing

Report Components (Typical Clinical)

• Identifying info
• Referral question
• Tests administered & dates
• Behavioral observations (interview & test-situation)
• Test results & interpretations (organized by construct)
• Summary & formulation
• Diagnostic impression (e.g., \text{DSM-5} code)
• Recommendations (treatment, accommodations)

Industrial/Organizational Variant

• ID info
• Test battery summary
• Skills & abilities profile
• Personality/job-fit profile
• Recommendations for placement/development

Writing Principles

• Validate statements with behavioral evidence or score data.
• Balance strengths & weaknesses (use “sandwich” method).
• Avoid hedging (“it appears…”) unless uncertainty must be conveyed.
• Respect confidentiality; use plain language if client to read report.