Comrey Personality Inventory

1. Abstract

The Comrey Personality Inventory (CPI; Comrey, 1964) represents a foundational milestone in psychometric taxonomy and the structural assessment of human personality. Constructed by renowned quantitative psychologist Andrew L. Comrey, the instrument was engineered to resolve critical psychometric shortcomings prevalent in contemporary multiscale inventories, such as excessive item heterogeneity, item-level unreliability, and ambiguous factor structures. The 1964 version of the inventory comprises 216 self-report items systematically configured into 36 distinct six-item subscales, operationalized psychometrically as Factored Homogeneous Item Dimensions (FHIDs). These scales encompass 33 substantive personality dimensions alongside three dedicated validity and response-bias scales: the Validity Scale (assessing random, irrational, or careless responding), the Truth Scale (quantifying candor and willingness to endorse unflattering self-truths), and the Social Desirability Scale (measuring the propensity toward normative self-falsification and impression management).

The CPI utilizes two balanced, nine-point response formats—Scale X (frequency: Never to Always) and Scale Y (probability/agreement: Absolutely Not to Absolutely)—which mitigate dichotomous forced-choice artifacts and reduce acquiescent response bias. Extensive factor analytic investigations across normative cohorts of university students and community adults demonstrated high internal consistency across the primary subscales, with a median reliability coefficient of .835 (individual dimensions ranging from .62 to .93). Higher-order and intermediate exploratory factor extractions rotated via Varimax and oblique analytic criteria confirmed robust overarching second-order factors, most notably Compulsion, Dependence, Hostility, and Neuroticism. Serving as the primary empirical progenitor to the formalized Comrey Personality Scales (CPS), the 1964 inventory remains a paramount exemplar of hierarchical scale construction, radical item homogeneity, and rigorous multivariate measurement theory in individual differences research.

2. Keywords

Comrey Personality Inventory, Andrew L. Comrey, Factored Homogeneous Item Dimensions, Psychometrics, Factor Analysis, Compulsion, Dependence, Hostility, Neuroticism, Social Desirability, Personality Assessment, Scale Construction

3. Authors

The Comrey Personality Inventory was conceived, validated, and published by Andrew Laurence Comrey, Ph.D. (1923–2011). Comrey was an eminent American quantitative psychologist, psychometrician, and longtime Professor of Psychology at the University of California, Los Angeles (UCLA). After completing his doctoral work under the supervision of psychometric pioneers, Comrey dedicated much of his academic career at UCLA to refining structural equation concepts, developing iterative minimum residual factor analysis algorithms, and eradicating measurement error in personality questionnaires.

Throughout the 1950s, 1960s, and 1970s, Comrey served as a principal authority in multivariate behavioral research, publishing landmark texts on factor analysis and psychometric design. Inquiries regarding historical archives, technical manuals, and archival psychometric research protocols associated with Comrey’s laboratory are curated within the institutional repositories of the UCLA Department of Psychology and affiliated academic measurement archives.

4. Purpose

The principal objective of the Comrey Personality Inventory (1964) was to overcome the structural instabilities, excessive item-level error variance, and factorial ambiguities that plagued earlier trait inventories, such as early iterations of the Minnesota Multiphasic Personality Inventory (MMPI) and the Guilford-Zimmerman Temperament Survey. Comrey posited that personality assessment instruments frequently failed replication attempts because individual questionnaire items are intrinsically noisy, susceptible to multiple interpretations, and burdened with substantial specific variance. Consequently, factor-analyzing raw item correlation matrices routinely yielded ephemeral pseudo-factors driven by item difficulty, linguistic overlap, or idiosyncratic response styles rather than authentic structural traits.

To rectify these psychometric vulnerabilities, the CPI was designed around the structural concept of Factored Homogeneous Item Dimensions (FHIDs). By grouping tightly related, highly focused items into six-item elementary mini-scales, Comrey sought to construct an instrument whose basic units of factor analysis possessed high internal consistency and minimal error variance. The primary clinical, organizational, and research objectives of the inventory include:

• Multi-Dimensional Trait Profiling: Systematically profiling an individual’s psychological functioning across 33 differentiated personality dimensions, providing an exhaustive evaluation of dispositional tendencies.
• Detection of Response Distortion: Identifying non-credible or distorted test protocols through three embedded validity indices that isolate careless responding (Validity), psychological defensiveness (Truth), and conscious or unconscious impression management (Social Desirability).
• Higher-Order Structural Mapping: Measuring core structural domains of adult personality—specifically Compulsion, Dependence, Hostility, and Neuroticism—which align with modern conceptualizations of psychopathology and normal personality variation.
• Applied Assessment: Serving as an objective diagnostic aid in counseling contexts, vocational guidance, academic advisement, and empirical research investigating the genetic, developmental, and behavioral correlates of personality traits.

5. Psychological Construct

The psychological constructs measured by the Comrey Personality Inventory span both primary-level behavioral tendencies and higher-order structural domains. At the core of Comrey’s measurement philosophy is the aggregation of behavior into 33 foundational personality dimensions, complemented by three structural response validity indicators. Each dimension is operationalized through six highly focused items that reflect explicit behavioral tendencies rather than vague introspective impressions.

Validity and Response Set Constructs

Comrey recognized that without rigorous control for response biases, substantive personality interpretations remain fundamentally confounded. The CPI incorporates three targeted control constructs:

• Validity (V): Designed to measure uncooperativeness, extreme carelessness, cognitive confusion, or outright bizarre responding. Items on this scale describe conditions that are virtually impossible or universally rejected by rational individuals; high scores indicate an invalid protocol attributable to random marking or reading comprehension failure.
• Truth (T): Measures an individual’s willingness to acknowledge minor, universal human shortcomings and unflattering personal truths. Modeled conceptually after the L (Lie) scale of clinical inventories, individuals with elevated Truth scores exhibit candid self-awareness, whereas depressed scores indicate defensiveness and denial of ubiquitous personal flaws.
• Social Desirability (S): Captures the deliberate or subconscious tendency to portray oneself in an exceptionally favorable, moral, and socially conventional light, obscuring genuine behavioral tendencies beneath culturally sanctioned platitudes.

Substantive Higher-Order and Primary Constructs

Through empirical factor analysis of the substantive dimensions, Comrey isolated four major macro-domains that account for the primary variance across the inventory:

• Compulsion: This construct captures individual differences in orderliness, precision, routine, meticulousness, and behavioral perfectionism. High scorers are punctilious, methodical, highly organized, and distressed by ambiguity or disarray. Lower scorers exhibit flexibility, tolerance for disorder, casualness regarding punctuality, and unstructured work styles.
• Dependence: Measures the psychological necessity for interpersonal support, external reassurance, affiliation, and guidance in decision-making. Elevated scores reflect reliance on others, fear of social isolation, and an externalized locus of emotional stability. Low scores signify autonomy, fierce self-reliance, and emotional detachment from peer approval.
• Hostility: Operationalizes interpersonal antagonism, irritability, resentfulness, readiness to engage in physical or verbal confrontation, and generalized distrust toward others. Low scores reflect empathy, cooperativeness, peaceful demeanor, and tolerance, while high scores indicate combative, vindictive, and reactive interpersonal tendencies.
• Neuroticism: Reflects chronic emotional instability, anxiety, depressive affect, somatic tension, feelings of inadequacy, and vulnerability to stress. High scorers experience frequent mood disruptions and perceived distress, whereas low scorers exhibit emotional resilience, calm demeanor, and subjective well-being.

Surrounding these core higher-order constructs are extensive secondary clusters measuring sociability, dominance, ascendance, empathy, and intellectual persistence, establishing a nuanced and comprehensive map of the human personality architecture.

6. Theoretical Framework

The Comrey Personality Inventory is grounded in the structural, quantitative tradition of trait psychology, heavily influenced by the psychometric theories of L. L. Thurstone, J. P. Guilford, and Raymond B. Cattell. Comrey approached personality not from a psychoanalytic or purely clinical perspective, but from an objective, empirical, taxonomic perspective. The inventory’s conceptual framework is governed by three foundational psychometric postulates: the Factored Homogeneous Item Dimension (FHID) hypothesis, the hierarchical structure of personality traits, and the necessity of metric standardization.

The Factored Homogeneous Item Dimension (FHID) Paradigm

During the 1950s and 1960s, academic psychometrics experienced severe friction between proponents of single-item factor analysis and proponents of broad clinical scales. Comrey argued that individual questionnaire items are psychometrically defective foundational units for factor extraction. A single test item possesses a low signal-to-noise ratio, is subject to idiosyncratic linguistic misinterpretations, and inherently yields low reliability coefficients (often ranging from .20 to .40). When hundreds of individual items are factored together, mathematical artifacts—such as “difficulty factors” and spurious correlations—inevitably distort the latent structure.

To solve this crisis, Comrey introduced the concept of the Factored Homogeneous Item Dimension. Under this paradigm, the psychometrician writes a small, tightly bounded cluster of items (exactly six in the CPI) that assess an identical, narrowly defined facet of behavior. These six items are carefully balanced, evaluated, and summed to produce a mini-scale or FHID score. Because the FHID consists of six parallel indicators, its internal consistency is significantly higher than that of an isolated item. Subsequent higher-order factor analyses are then conducted on the covariance matrix of these homogeneous dimensions rather than raw items, ensuring superior mathematical stability, replicability, and clean structural extraction.

Hierarchical Trait Organization

Comrey’s theoretical framework adheres strictly to a hierarchical model of behavioral organization. At the lowest level of the hierarchy are specific observable behaviors or item responses. At the second level are the FHIDs, which aggregate multiple instances of congruent behaviors into coherent, narrow-band habits. At the third level are the intermediate factors revealed through the intercorrelations among the FHIDs. At the apex of the hierarchy are the broad, generalized higher-order personality dimensions—such as Compulsion, Dependence, Hostility, and Neuroticism—which explain broad swaths of human behavioral variance across diverse ecological contexts. This model presaged modern structural taxonomies, bridging the gap between narrow facet-level inventories and broad five-factor personality paradigms.

7. Validity

The validation of the Comrey Personality Inventory was executed through systematic empirical procedures designed to demonstrate content, construct, convergent, and discriminant validity across distinct adult populations.

Construct and Factorial Validity

Construct validity in the CPI was fundamentally embedded into its test construction protocol. Comrey utilized an iterative factor-analytic purification process. Across initial normative validation samples consisting of hundreds of university students and community residents, Comrey demonstrated that the items assigned to each FHID loaded heavily on their targeted primary factor while maintaining negligible cross-loadings on extraneous dimensions. The construct validity of the core dimensions—Compulsion, Dependence, Hostility, and Neuroticism—was confirmed by replicating their factor extraction across sex-stratified and demographically distinct cohorts, demonstrating structural invariance.

Convergent and Discriminant Validity

Subsequent psychometric evaluations examined the convergence between Comrey’s extracted factors and external personality benchmarks, including the Guilford-Zimmerman Temperament Survey, Cattell’s Sixteen Personality Factor Questionnaire (16PF), and the MMPI. Empirical findings confirmed strong patterns of convergent validity:

• The CPI Neuroticism factor correlated strongly with the MMPI Psychasthenia (Pt), Depression (D), and Taylor Manifest Anxiety scales ($r > .60$), demonstrating robust convergence in assessing emotional maladjustment, while remaining largely orthogonal to cognitive ability and vocational aptitude measures.
• The Hostility dimension exhibited statistically significant positive correlations with external measures of interpersonal aggression, trait anger, and overt antagonism, alongside inverse correlations with measures of social cooperativeness and friendliness ($r = -.50$ to $-.65$).
• The Compulsion factor demonstrated precise convergence with external scales assessing orderliness, ritualistic behavior, and planning persistence, while establishing clear discriminant validity against affective and anxiety-driven indices.

Validity Scales Functionality

The empirical effectiveness of the Validity, Truth, and Social Desirability scales was verified through experimental “fake good” and “fake bad” instructional sets. When participants were instructed to simulate idealized job candidates, scores on the Social Desirability scale demonstrated marked, statistically significant increases ($p < .001$), while Truth scores plummeted. Conversely, under instructions to feign severe psychopathology or respond randomly, the Validity scale exhibited sharp elevations, successfully identifying over 95% of non-genuine response protocols and confirming the utility of these control metrics in high-stakes assessment.

8. Reliability

The reliability architecture of the Comrey Personality Inventory was rigorously documented by Comrey (1964) utilizing both classical internal consistency methodologies and test-retest stability coefficients.

Internal Consistency Reliability

Because each of the 30 substantive personality dimensions (and the three response control scales) was structured as a six-item homogeneous unit (FHID), Comrey calculated internal consistency coefficients across large normative cohorts comprising university undergraduates and adult community residents. The psychometric results yielded:

• Substantive Personality Dimensions: Across the 30 primary personality subscales, the median internal consistency coefficient was .835, with individual subscale values ranging from a low of .62 to an exceptional high of .93. These figures are remarkably high considering that each subscale is composed of only six individual items.
• Response Validity Scales: The Truth Scale demonstrated an internal consistency reliability of .69, reflecting the somewhat broader behavioral spectrum associated with acknowledging varied personal shortcomings. The Social Desirability Scale yielded a reliability coefficient of .58, an expected finding given the multidimensional, heterogeneous nature of impression management behaviors.

Composite and Test-Retest Stability

When the homogeneous subscales were aggregated into their respective higher-order factors (Compulsion, Dependence, Hostility, and Neuroticism), the resulting macro-scale internal consistencies consistently exceeded .90, demonstrating outstanding measurement precision. Furthermore, short-term and medium-term test-retest reliability investigations spanning intervals of two to six weeks demonstrated stability coefficients ranging from .75 to .88 across the primary dimensions, confirming that the CPI measures enduring dispositional traits rather than transient psychological states.

9. Factor Analysis

The statistical core of the Comrey Personality Inventory rests upon exhaustive, multi-stage factor analytic investigations conducted by Comrey in the early 1960s. Rather than relying on standard commercial factoring packages of the era, Comrey utilized custom mathematical algorithms, specifically applying iterative minimum residual factor analysis and rigorous structural rotations to establish the latent dimensionality of the inventory.

Factor Extraction and Rotation Procedures

In the primary 1964 investigation, Comrey gathered item and dimension scores from a diverse sample of university students and community-dwelling adults. The data analysis proceeded through several structured phases:

1. Correlation Matrix Computation: Product-moment correlation matrices were generated based on total scores across the factored homogeneous item dimensions.
2. Initial Factor Extraction: A total of 25 factors were extracted from the initial unrotated correlation matrix using principal axis / minimum residual techniques to account for common variance while excluding unique error components.
3. Orthogonal Rotation: The initial 25 factors were subjected to an orthogonal transformation using the Varimax criterion to identify salient clusters of variance and clean linear separations.
4. Oblique Simple Structure Transformation: Comrey isolated the 17 factors that exhibited meaningful, substantial loadings—defined by a threshold criterion of .30 or higher on primary variables. These 17 robust factors were subsequently rotated to an oblique simple structure, permitting moderate correlations among latent dimensions to mirror natural psychological covariation.

Higher-Order Factor Structures

The secondary and higher-order factor solutions yielded clean, interpretable configurations that clarified longstanding debates in personality psychology. Most notably, four preeminent second-order dimensions consolidated the primary FHIDs:

This mathematical demonstration proved that large batteries of personality items could be systematically reduced to stable, replicated structural dimensions without forfeiting the fine-grained variance captured by narrow-band homogeneous subscales.

10. Instrument / Measurement Tool

The Comrey Personality Inventory is a standardized, self-report psychological questionnaire. Below are its structural and operational specifications:

• Test Type: Multi-scale objective personality inventory / self-report psychometric questionnaire.
• Year of Development: 1964.
• Item Count: 216 total items.
• Scale Structure: 36 distinct scales (comprising 33 substantive personality dimensions and 3 response-validity scales). Each scale consists of exactly 6 items.
• Administration Modality: Paper-and-pencil questionnaire administration; group or individual format.
• Target Population: Adults and adolescents aged 16 and older (human males and females; normative data derived from university students and community resident samples).
• Administration Time: Approximately 35 to 50 minutes.
• Response Format: Two differentiated nine-point rating formats engineered to prevent response flattening:
• Scale X (Frequency Scale): A 9-point continuum anchored from 1 = Never to 9 = Always (evaluates behavioral frequency).
• Scale Y (Probability/Agreement Scale): A 9-point continuum anchored from 1 = Absolutely Not to 9 = Absolutely (evaluates subjective certainty and attitudinal agreement).
• Scoring Methodology: Subscale raw scores are calculated by summing the six items within each FHID (after reversing negatively phrased items). FHID scores range from 6 to 54. Primary dimensions are subsequently aggregated or profiled against standardized normative tables (e.g., T-scores or stanines) to generate composite factor indices for Compulsion, Dependence, Hostility, Neuroticism, and associated personality clusters.

11. Permissions & Fee and Test Year

The Comrey Personality Inventory was published in 1964 by Andrew L. Comrey as part of empirical scale development studies released in academic literature. The technical psychometrics and initial structural design were detailed in the peer-reviewed journal Educational and Psychological Measurement (Comrey, 1964). The instrument served as the direct precursor to the commercially distributed Comrey Personality Scales (CPS), later published in 1970 by the Educational and Industrial Testing Service (EdITS).

For historical research, non-commercial academic investigations, and educational teaching purposes, the general measurement architecture and empirical findings published in Comrey’s 1964 scholarly articles may be utilized in accordance with academic fair use. However, the complete, verbatim copyrighted test items, commercial scoring keys, and copyrighted testing booklets for modern derivative instruments (CPS) remain the intellectual property of the publisher (EdITS / Educational and Industrial Testing Service). Researchers intending to administer the official copyrighted tool in clinical or organizational settings must obtain appropriate authorization, licensing, and testing materials directly from the copyright holder.

12. References

The following peer-reviewed publications and archival sources document the empirical development, factor analysis, and validation of the Comrey Personality Inventory:

• Comrey, A. L. (1961). Factored homogeneous item dimensions in personality research. Educational and Psychological Measurement, 21(2), 417–431. https://doi.org/10.1177/001316446102100213
• Comrey, A. L. (1962). A study of thirty-five personality dimensions. Educational and Psychological Measurement, 22(3), 543–552. https://doi.org/10.1177/001316446202200310
• Comrey, A. L. (1964). Personality factors compulsion, dependence, hostility, and neuroticism. Educational and Psychological Measurement, 24(1), 75–84. https://doi.org/10.1177/001316446402400106
• Comrey, A. L. (1970). EdITS manual for the Comrey Personality Scales. Educational and Industrial Testing Service.
• Comrey, A. L., & Lee, H. B. (1992). A first course in factor analysis (2nd ed.). Lawrence Erlbaum Associates.
• Guilford, J. P. (1959). Personality. McGraw-Hill.
• Thurstone, L. L. (1947). Multiple factor analysis. University of Chicago Press.

13. Items of the Scale