Physical Symptoms Inventory (PSI)

Abstract

The Physical Symptoms Inventory (PSI), developed by Paul E. Spector and Steve M. Jex in 1998, is a widely recognized psychometric assessment tool designed to quantify self-reported somatic health complaints and physiological strain within occupational, clinical, and organizational research settings. Consisting primarily of an 18-item primary version alongside refined 13-item and 12-item variants, the instrument captures a diverse array of somatic manifestations, including gastrointestinal disturbances, musculoskeletal pain, cardiorespiratory symptoms, sleep disruption, and general autonomic or neurological distress experienced over a designated recall period (typically the preceding 30 days). Respondents rate the frequency of each symptom using a 5-point Likert-type scale ranging from 1 ("Not at all") to 5 ("Every day"), or alternatively via a binary occurrence metric or a healthcare-utilization categorization. Grounded within occupational stressor-strain paradigms and psychoneuroimmunological models of chronic stress, the PSI serves as a vital bridge linking subjective psychosocial workplace stressors—such as interpersonal conflict, heavy workload, and organizational constraints—to tangible bodily manifestations. Psychometrically, the instrument demonstrates robust internal consistency (Cronbach's α typically spanning .75 to .86 across heterogeneous organizational samples) and adequate test-retest stability. Crucially, the authors and subsequent psychometricians emphasize that the scale functions effectively as a formative or causal indicator composite, rather than an exclusively reflective latent construct, in accordance with measurement principles articulated by Kenneth Bollen and Richard Lennox. Extensive construct, convergent, and criterion-related validity analyses consistently show that the PSI correlates significantly with organizational stressors, affective strain (e.g., anxiety, depression, burnout), negative affectivity, sickness absenteeism, and reduced productivity, establishing it as an essential measurement standard in occupational health psychology.

Keywords

Physical Symptoms Inventory, PSI, somatic strain, occupational stress, psychosomatic complaints, job strain, Spector and Jex, health outcomes, organizational health psychology, formative measurement, psychometrics.

Authors

The Physical Symptoms Inventory was developed by two prominent scholars in the field of industrial-organizational psychology and occupational health:

• Paul E. Spector, Ph.D.: Distinguished Professor Emeritus of Industrial-Organizational Psychology at the University of South Florida. Dr. Spector is an internationally renowned psychometrician and organizational scholar recognized for developing numerous widely utilized measurement scales, including the Job Satisfaction Survey (JSS), the Interpersonal Conflict at Work Scale (ICAWS), the Organizational Constraints Scale (OCS), and the Counterproductive Work Behavior Checklist (CWB-C). His research interests center on job stress, occupational health psychology, workplace aggression, and survey methodology.
• Steve M. Jex, Ph.D.: Professor of Psychology and former Director of the Industrial-Organizational Psychology Doctoral Program at University of Central Florida and Bowling Green State University. Dr. Jex's pioneering work investigates organizational stressors, employee well-being, the physiological correlates of occupational strain, workplace incivility, and interventions designed to foster employee resilience.

Direct inquiries, research permissions, and instrument archives are managed through Dr. Paul E. Spector's academic repository hosted by the University of South Florida.

Purpose

The primary purpose of the Physical Symptoms Inventory is to provide an efficient, standardized, and psychometrically defensible instrument for assessing the somatic outcomes of stress. While psychological strain constructs—such as emotional exhaustion, state-trait anxiety, depressive symptomatology, and diminished job satisfaction—have historically dominated organizational and epidemiological research, somatic strain represents a critical downstream consequence of prolonged psychological distress. Physical symptoms provide concrete, physiological evidence of the human toll exacted by adverse workplace environments, severe life events, and ongoing cognitive and emotional overload.

Historically, researchers faced significant methodological challenges when assessing physical health via self-report. Many existing medical inventories were excessively long, clinically oriented toward diagnosing specific medical conditions (e.g., DSM somatization disorder), heavily confounded with preexisting chronic medical conditions, or contaminated by subjective emotional evaluations. Spector and Jex (1998) addressed these limitations by constructing an inventory focused strictly on functional somatic symptoms commonly triggered or exacerbated by sympathetic nervous system activation and chronic allostatic load, rather than complex systemic medical diagnoses.

In occupational health psychology, human resource management, and epidemiology, the PSI serves several essential research and practical functions:

• Quantifying Somatic Strain: It acts as a standardized criterion variable when testing transactional and structural models of occupational stress, such as the Job Demands-Resources (JD-R) model and the Demand-Control-Support model.
• Organizational Health Audits: It provides occupational health practitioners and corporate wellness professionals with benchmark data to identify high-strain organizational departments, assess the systemic burden of workplace dysfunction, and evaluate ergonomic or psychosocial interventions.
• Clinical and Screening Utility: While not designed as a definitive diagnostic instrument for specific pathology, high aggregate scores on the PSI indicate generalized bodily distress, somatization, and autonomic dysregulation, serving as an early indicator of employee burnout, impending medical absenteeism, or chronic illness onset.
• Controlling for Contaminants: Shortened variants (the 13-item and 12-item versions) deliberately excise symptoms likely to stem from infectious, non-stress-related physiological etiologies (e.g., fever, skin rashes, infections), allowing researchers to isolate somatic symptoms directly linked to psychoneuroendocrine stress pathways.

Psychological Construct

The focal construct measured by the Physical Symptoms Inventory is somatic strain (frequently referred to in the literature as physical strain, psychosomatic complaints, or subjective bodily distress). Within occupational health psychology, strain is conceptualized as the multifaceted physiological, psychological, and behavioral response of an individual to prolonged or intense environmental demands and stressors.

Somatic strain specifically reflects subjective physical discomfort and functional somatic symptoms that manifest when an organism's coping resources are overwhelmed. Unlike clinical diagnoses that denote structural organic pathology, somatic symptoms often reflect functional dysregulations across major physiological systems:

• Gastrointestinal Distress: Symptoms such as upset stomach, nausea, acid indigestion, heartburn, stomach cramps, diarrhea, and constipation. These reactions are directly mediated by stress-induced alterations in the brain-gut axis, delayed gastric emptying, and disrupted gastrointestinal motility prompted by corticotropin-releasing factor (CRF) and sympathoadrenal activation.
• Musculoskeletal Tension: Manifested as backaches and generalized muscular strain. Prolonged isometric contraction of skeletal muscles, particularly in the trapezius and lumbar regions, occurs as a persistent neuromuscular response to environmental threat or cognitive vigilance.
• Cardiorespiratory and Autonomic Arousal: Illustrated by chest pain, shortness of breath, heart pounding in the absence of physical exercise, and dizziness. These symptoms indicate sudden shifts in vagal tone, hyperventilation, and surges in circulating catecholamines (epinephrine and norepinephrine).
• Sleep Disruption and Chronic Fatigue: Items capturing trouble sleeping and persistent tiredness or fatigue measure dysregulation of the circadian hypothalamic-pituitary-adrenal axis, fragmented sleep architecture, and insufficient physiological recovery.
• Neurological and Sensory Complaints: Manifested through tension headaches, eye strain, and tinnitus (ringing in the ears), frequently exacerbated by continuous cognitive demands, mental fatigue, and sustained micro-arousal.

A central psychometric consideration highlighted by Spector and Jex concerns whether somatic strain constitutes a reflective latent construct or a formative composite. In classical reflective measurement models, an underlying latent trait directly causes the observable manifestations, necessitating high inter-item correlations. Conversely, as posited by Bollen and Lennox (1991), the PSI can be effectively conceptualized as a formative index: an individual can experience severe chronic backaches and insomnia without experiencing gastrointestinal cramps or skin rashes. Each item contributes distinct variance to an individual's cumulative somatic burden, meaning that low inter-item correlations do not compromise the construct validity of the aggregate index.

Theoretical Framework

The Physical Symptoms Inventory is anchored in several foundational theories of stress, psychophysiology, and occupational health.

1. The Job Stressor-Strain Model

Originating from the work of scholars such as Robert Kahn, Daniel Katz, and later expanded by Cary Cooper and Paul Spector, the Job Stressor-Strain Model posits that objective and subjective environmental conditions (stressors) elicit negative internal states in individuals (strain). Physical symptoms represent the biological component of the strain response. When workers encounter persistent stressors—such as extreme quantitative workload, organizational constraints, or interpersonal hostility—their cognitive appraisal initiates a systemic neurobiological cascade that, if unresolved, produces observable somatic complaints.

2. The Transactional Model of Stress and Coping

Formulated by Richard Lazarus and Susan Folkman (1984), this framework emphasizes that stress arises from an ongoing transaction between the person and the environment. When environmental demands are appraised as exceeding personal coping resources, distress ensues. The PSI captures the biological cost of coping mechanisms that have proven inadequate to neutralize chronic stressors, reflecting persistent states of physiological mobilization.

3. Allostatic Load Theory

Developed by neuroendocrinologist Bruce McEwen (1998), the concept of allostasis describes the active process through which the body maintains physiological stability through change. While acute allostatic responses are adaptive for survival, prolonged or unmitigated activation of the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic-adrenomedullary (SAM) axis results in allostatic load. Over time, excessive glucocorticoid circulation, sustained elevated blood pressure, and altered immune responses cause wear and tear on organ systems, manifesting as the somatic symptoms inventoried by the PSI (e.g., gastrointestinal mucosal breakdown, cardiovascular palpitations, and muscular tension).

4. Conservation of Resources (COR) Theory

Stevan Hobfoll's (1989) Conservation of Resources theory suggests that psychological stress occurs when individuals face resource loss, threat of resource loss, or a failure to gain resources following significant investment. Energetic and physical capacities represent fundamental personal resources. When an individual persistently depletes their physiological stamina without adequate replenishment, somatic symptoms emerge as a direct indicator of physiological resource depletion.

Validity

The validity of the Physical Symptoms Inventory has been extensively established across diverse occupational sectors, international cohorts, and clinical studies.

Construct and Convergent Validity

In the seminal validation study by Spector and Jex (1998), the PSI was administered across six independent employee samples comprising 1,223 total participants, including civil service employees, university administrative staff, healthcare workers, and retail personnel. The authors evaluated convergent validity by correlating PSI scores with three established occupational stressor inventories:

• Interpersonal Conflict at Work Scale (ICAWS): Correlations between somatic symptoms and interpersonal conflict ranged from .23 to .41 (p < .01), demonstrating that social friction in the workplace consistently manifests in bodily distress.
• Organizational Constraints Scale (OCS): Correlations between situational barriers to task performance and the PSI ranged from .27 to .43 (p < .01), indicating that frustration stemming from workplace obstacles yields direct physiological strains.
• Quantitative Workload Inventory (QWI): Associations with excessive pacing and volume of work ranged between .18 and .37 (p < .01).

Furthermore, convergent validity is evidenced by moderate to strong correlations with well-established measures of psychological distress, including the General Health Questionnaire (GHQ), the Beck Depression Inventory (BDI), and the Maslach Burnout Inventory (MBI), particularly the exhaustion subscale (typical correlations r = .40 to .55).

Discriminant Validity

A recurring methodological issue in self-report health research is whether physical symptom reporting is merely an artifact of Negative Affectivity (NA) or neuroticism. Spector and Jex examined whether the relationship between job stressors and PSI scores remained significant after partialling out negative affectivity. While NA correlates with both stressors and somatic complaints, multiple hierarchical regression analyses confirm that workplace stressors account for significant incremental variance in PSI scores (often explaining an additional 5% to 12% of unique variance) above and beyond the effects of dispositional negative affectivity. Moreover, discriminant validity analyses demonstrate that the PSI can be statistically differentiated from organizational attitude measures such as affective organizational commitment and job satisfaction.

Criterion-Related and Predictive Validity

The PSI demonstrates substantial predictive utility with respect to objective organizational outcomes:

• Sickness Absenteeism: Longitudinal studies demonstrate that baseline PSI scores reliably predict future certified sick leave days, incidental absences, and physician consultations over 6-month and 12-month follow-up intervals (coefficients ranging from .22 to .38).
• Turnover Intentions and Actual Turnover: Elevated physical symptom reporting serves as an intermediate strain variable predicting voluntary organizational exit.
• Presenteeism: Recent epidemiological research indicates that employees reporting elevated somatic complaints on the PSI exhibit significant performance decrements, cognitive errors, and decreased contextual performance while on the job.

Reliability

Psychometric evaluations of the Physical Symptoms Inventory show satisfactory to high reliability estimates across numerous empirical studies.

Internal Consistency

In the initial development paper by Spector and Jex (1998), Cronbach's alpha internal consistency estimates across five distinct samples ranged from .75 to .85 for the 18-item scale (mean α ≈ .80). For the shortened 13-item and 12-item versions, internal consistency estimates similarly range between .74 and .84.

Measurement Model Considerations: Spector and Jex (1998) and subsequent psychometric scholars explicitly observe that when interpreting Cronbach's alpha for the PSI, classical test theory assumptions must be evaluated carefully. Because the inventory functions largely as a formative index of diverse physiological symptoms across disparate biological systems (e.g., a person experiencing eye strain and backaches may not necessarily experience diarrhea), the internal consistency of such scales can underestimate overall reliability. Despite this theoretical constraint, the PSI routinely achieves alpha levels comfortably exceeding the standard .70 psychometric threshold established by Nunnally and Bernstein.

Test-Retest Stability

Temporal stability assessments have demonstrated moderate to strong test-retest reliability across various time horizons. Over short intervals (e.g., 2 to 4 weeks), test-retest correlations typically fall within the .70 to .82 range, indicating that individuals report symptom frequencies with consistent temporal reliability. Over longer periods (6 to 12 months), stability coefficients typically moderate to the .45 to .60 range. This expected decline in long-term stability confirms that the PSI is sensitive to dynamic changes in environmental stressors and coping adaptations, reflecting an episodic physiological strain state rather than an invariant personality trait.

Factor Analysis

The structural dimensionality of the Physical Symptoms Inventory has been thoroughly investigated through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Structure

Early exploratory factor analyses conducted on the 18-item PSI generally yield a strong general first factor representing generalized somatic discomfort, which accounts for the largest proportion of total variance (typically between 28% and 35%). When unconstrained multi-factor extraction methods (such as principal axis factoring with promax or varimax rotation) are applied, items frequently load onto distinct physiological symptom clusters:

1. Gastrointestinal Factor: High loadings (> .55) from upset stomach/nausea, acid indigestion/heartburn, diarrhea, stomach cramps, and constipation.
2. Musculoskeletal and Pain Factor: Primary loadings from backache, headache, and eye strain.
3. Cardiorespiratory and Neurological Arousal: Substantial loadings from heart pounding when not exercising, chest pain, shortness of breath, and dizziness.
4. Asthenia/Fatigue and Sleep Disruption: Loadings dominated by trouble sleeping, tiredness/fatigue, and loss of appetite.

Confirmatory Factor Analysis and Structural Models

Confirmatory factor analytic investigations evaluating the PSI have compared unidimensional, hierarchical, and multi-factor structures. While a four-factor or five-factor correlated model often provides a slightly superior absolute empirical fit (χ²/df < 2.5, CFI > .92, RMSEA ≤ .055), these underlying factors are typically highly intercorrelated (rs ranging from .50 to .75). Consequently, Spector and Jex advocated for treating the scale as a predominantly unidimensional composite in practical and organizational research, noting that a single composite score effectively captures cumulative somatic strain.

Formative vs. Reflective Model Considerations

In structural equation modeling (SEM) applications, researchers frequently follow the recommendations of Bollen and Lennox (1991) by modeling the PSI as a formative measurement model (cause indicators) rather than a reflective latent variable (effect indicators). Under a formative specification, the composite somatic strain construct is defined as a linear combination of the individual symptoms. This approach eliminates the statistical assumption that all bodily symptoms must covary strongly with one another, providing an accurate representation of how heterogeneous somatic symptoms coalesce to reflect an individual's overall physical strain level.

Instrument / Measurement Tool

• Test Name: Physical Symptoms Inventory (PSI)
• Alternative Titles: Physical Strain Questionnaire; Spector & Jex Physical Symptoms Scale
• Developers: Paul E. Spector, Ph.D., and Steve M. Jex, Ph.D.
• Year of Development: 1998
• Construct Assessed: Self-reported somatic strain and psychosomatic physical symptoms resulting from environmental stress
• Administration Format: Paper-and-pencil questionnaire, online assessment, or integrated psychological survey module
• Completion Time: Approximately 3 to 5 minutes
• Target Population: Working adults, university students, and general community populations (ages 18 and older)
• Versions Available:
  • 18-Item Version: Full primary inventory assessing the broad spectrum of physical complaints over the past 30 days.
  • 13-Item Version: Refined variant excluding non-stress-related symptoms (e.g., skin rash, fever, infection, chest pain, shortness of breath) and adding "Ringing in the ears".
  • 12-Item Version: Abbreviated version identical to the 13-item scale but omitting "A backache" to reduce contamination from physical/ergonomic labor.
• Response Scale Options:
  • Standard 5-Point Frequency Scale (Mandatory Primary Format):
    • 1 = Not at all
    • 2 = Once or twice
    • 3 = Once or twice per week
    • 4 = Several times per week
    • 5 = Every day
  • Binary Format: 0 = No, 1 = Yes
  • Healthcare Utilization Format:
    • 1 = No
    • 2 = Yes, but I didn't see doctor
    • 3 = Yes, and I saw doctor
• Scoring and Interpretation Procedures:
  • Scoring Formula: Total scores can be calculated either as a simple sum of item scores (ranging from 18 to 90 for the 18-item version using the 1–5 scale) or as a mean item response score (ranging from 1.0 to 5.0).
  • Reverse-Scoring: None. All items are positively keyed toward symptomatic distress; higher aggregate scores unequivocally indicate greater somatic strain.
  • Shorter Version Rationale: Researchers investigating purely psychosocial or cognitive stressors often utilize the 13-item or 12-item versions to eliminate somatic symptoms that frequently stem from contagious pathogens, biological infections, or primary ergonomic strain.

Permissions & Fee and Test Year

The Physical Symptoms Inventory was initially published in 1998 in the Journal of Occupational Health Psychology by Paul E. Spector and Steve M. Jex. In alignment with Dr. Paul E. Spector's commitment to advancing scientific research in occupational health psychology, the PSI is widely accessible to the global scientific community.

• Permissions and Academic Use: The scale is free of charge for non-commercial research, academic, and educational purposes. Researchers and educational practitioners are permitted to reproduce and administer the scale in their investigations without seeking explicit individual written permission, provided that full and accurate academic citation is given to the original authors (Spector & Jex, 1998).
• Commercial Applications: Commercial organizations, management consulting firms, or vendors seeking to incorporate the PSI into commercial diagnostic platforms, revenue-generating assessment suites, or proprietary organizational audits must obtain formal permission from the authors or the respective copyright holders.
• Official Repository: Additional documentation, related stress scales (such as the ICAWS, OCS, and QWI), scoring instructions, and norms are hosted on Dr. Paul Spector's academic website at the University of South Florida.

References

• Bollen, K., & Lennox, R. (1991). Conventional wisdom on measurement: A structural equation perspective. Psychological Bulletin, 110(2), 305–314. https://doi.org/10.1037/0033-2909.110.2.305
• Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing stress. American Psychologist, 44(3), 513–524. https://doi.org/10.1037/0003-066X.44.3.513
• Keenan, A., & Newton, T. J. (1985). Stressful events, stressors, and psychological strains in young professional engineers. Journal of Occupational Behavior, 6(2), 151–156. https://doi.org/10.1002/job.4030060206
• Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer Publishing Company.
• McEwen, B. S. (1998). Stress, adaptation, and disease: Allostasis and allostatic load. Annals of the New York Academy of Sciences, 840(1), 33–44. https://doi.org/10.1111/j.1749-6632.1998.tb09546.x
• Peters, L. H., & O'Connor, E. J. (1980). Situational constraints and work outcomes: The influences of a frequently overlooked construct. Academy of Management Review, 5(3), 391–397. https://doi.org/10.5465/amr.1980.4288856
• Spector, P. E., & Jex, S. M. (1998). Development of four self-report measures of job stressors and strain: Interpersonal Conflict at Work Scale, Organizational Constraints Scale, Quantitative Workload Inventory, and Physical Symptoms Inventory. Journal of Occupational Health Psychology, 3(4), 356–367. https://doi.org/10.1037/1076-8998.3.4.356

Items of the Scale