Affect Balance Scale (ABS)

1. Abstract

The Affect Balance Scale (ABS), formulated by social psychologist Norman M. Bradburn in 1969, stands as one of the most seminal instruments in the history of subjective well-being, psychometrics, and psychiatric epidemiology. Designed to evaluate the psychological well-being of individuals within non-institutionalized general populations, the instrument broke fundamentally from the traditional psychiatric assumption that mental health exists on a unidimensional continuum running from profound distress to optimal wellness. Instead, Bradburn conceptualized subjective well-being as a dynamic equilibrium resulting from the interplay between two distinct, statistically independent dimensions: Positive Affect (PA) and Negative Affect (NA). The standard ABS consists of a 10-item self-report questionnaire utilizing a dichotomous (Yes / No) response format, asking respondents whether they have experienced specific emotional states over the preceding “past few weeks.” Five items measure positive affective states (e.g., excitement, pride, accomplishment), while five items measure negative affective states (e.g., restlessness, loneliness, depression). The overall Affect Balance score is derived by subtracting the Negative Affect sum from the Positive Affect sum, yielding a score reflecting the net affective equilibrium. Extensive psychometric evaluations across five decades have revealed modest to moderate internal consistency reliability (Cronbach’s alpha typically ranging from .55 to .73 for the independent subscales, reflecting the intentional heterogeneity of the brief item indicators) alongside strong test-retest reliability across brief assessment intervals (.70 to .80). Factor analytic investigations, both exploratory and confirmatory, have repeatedly affirmed the two-factor orthogonal structure, establishing the ABS as a foundational precursor to modern dual-process affective models and contemporary instruments such as the Positive and Negative Affect Schedule (PANAS). This article provides a comprehensive academic review of the scale’s historical context, theoretical foundations, psychometric architecture, validity evidence, and continued clinical and epidemiological utility.

2. Keywords

Affect Balance Scale, Norman Bradburn, subjective well-being, positive affect, negative affect, hedonic balance, psychological well-being, psychometrics, mental health assessment, emotional equilibrium

3. Authors

The Affect Balance Scale was conceptualized, developed, and empirically validated by Norman M. Bradburn, Ph.D., an eminent American social psychologist and methodologist associated with the National Opinion Research Center (NORC) and the University of Chicago. Bradburn’s seminal investigations into community psychological well-being began in collaboration with sociologist David Caplovitz in the mid-1960s, culminating in their joint pilot monograph, Reports on Happiness: A Pilot Study of Behavior Related to Mental Health (1965). Bradburn subsequently formalized the theoretical architecture, standardized the psychometric scoring, and conducted massive cross-sectional and longitudinal population-based validations, which were published in his landmark 1969 volume, The Structure of Psychological Well-Being.

Throughout his distinguished career, Dr. Bradburn served as the Tiffany and Margaret Blake Distinguished Service Professor Emeritus at the University of Chicago, holding joint appointments in the Department of Psychology, the Harris School of Public Policy, and the Booth School of Business. He also directed the National Opinion Research Center (1967–1973, 1979–1983) and served as the Assistant Director for Social, Behavioral, and Economic Sciences at the National Science Foundation (NSF). Inquiries regarding the historical development and archival survey data of the Affect Balance Scale are typically addressed through NORC at the University of Chicago (55 East Monroe Street, 30th Floor, Chicago, IL 60603, USA).

4. Purpose

The primary purpose of the Affect Balance Scale is the rapid, non-invasive measurement of overall psychological well-being and affective equilibrium in general population samples, clinical cohorts, and specialized demographic groups (such as older adults and chronic care populations). Prior to the late 1960s, clinical psychology and psychiatric epidemiology operated almost exclusively under a disease-model paradigm. Mental health was conceptualized as the mere absence of psychiatric symptomatology, such as anxiety, somatization, and clinical depression. Instruments developed during that era, including the Minnesota Multiphasic Personality Inventory (MMPI) and early screening inventories for neuroses, focused exclusively on psychopathology.

Bradburn developed the ABS to address a critical theoretical and methodological void: the inability of existing instruments to measure positive psychological states and holistic happiness. Bradburn argued that a comprehensive science of mental health must evaluate not only what makes individuals ill, but what makes their lives satisfying, enriching, and emotionally stable. The instrument was designed to achieve three foundational objectives:

• Population-Level Well-Being Surveillance: To serve as a standardized, brief, and easily administrable metric within large-scale demographic and epidemiological surveys to monitor population mental health across diverse socioeconomic, racial, and geographic strata.
• Testing the Orthogonality of Affect: To provide an empirical mechanism for measuring positive and negative emotional states simultaneously, allowing researchers to evaluate whether positive and negative affective experiences fluctuate inversely as polar opposites or operate as independent psychological dimensions.
• Assessing Overall Hedonic Balance: To calculate a single summary index—the affect balance—that reflects an individual’s overall net psychological state, serving as a reliable proxy for global subjective well-being, life satisfaction, and capacity to cope with environmental stressors.

In applied research, the ABS has been utilized extensively in gerontology to evaluate adaptation to retirement, bereavement, and institutionalization; in sociological studies assessing the psychological impacts of unemployment, urban density, and poverty; in behavioral medicine to examine affective determinants of physical morbidity, cardiovascular outcomes, and mortality; and in evaluation research to gauge the psychosocial impact of community-level interventions.

5. Psychological Construct

The core psychological construct measured by the Affect Balance Scale is Subjective Well-Being (SWB), specifically its affective or emotional component. Bradburn postulated that an individual’s immediate sense of happiness or life satisfaction is a function of the comparative ratio or balance between two distinct, qualitatively divergent experiential dimensions: Positive Affect and Negative Affect.

Positive Affect (PA)

Positive Affect refers to the extent to which an individual experiences pleasurable engagement, enthusiasm, psychological reward, pride, and active interest in their environment. Within the ABS framework, PA is not viewed merely as the absence of dysphoria, but as an active, appetitive psychological state driven by social engagement, goal attainment, novel stimulation, and environmental mastery. The five items comprising the Positive Affect subscale assess:

• Particularly excited or interested in something: Captures cognitive and motivational arousal, curiosity, and vitality.
• Proud because someone complimented you on something you had done: Reflects external social validation, self-worth, and relational recognition.
• Pleased about having accomplished something: Evaluates intrinsic mastery, self-efficacy, and task-related achievement.
• On top of the world: Represents peak affective experiences, exuberance, and intense joy.
• That things were going your way: Assesses environmental optimism, perceived serendipity, and external alignment with personal goals.

Negative Affect (NA)

Negative Affect represents an individual’s exposure to unpleasurable engagement, emotional distress, social isolation, and general psychological discomfort. Bradburn conceptualized NA as a protective, aversive emotional reaction to environmental frustration, loss of social connection, or structural barriers. The five items of the Negative Affect subscale measure distinct facets of psychological distress:

• So restless that you couldn’t sit long in a chair: Assesses psychomotor agitation, somatic tension, and generalized anxiety.
• Very lonely or remote from other people: Measures social estrangement, perceived alienation, and interpersonal deficits.
• Bored: Evaluates anhedonic disengagement, lack of environmental stimulation, and attentional apathy.
• Depressed or very unhappy: Probes core dysphoric mood, sorrow, and depressive symptomatology.
• Upset because someone criticized you: Gauges interpersonal vulnerability, emotional fragility, and negative social feedback reactivity.

The Affect Balance Model

Crucially, Bradburn’s construct posits that an individual’s net psychological well-being is not defined by either dimension in isolation, but by their mathematical discrepancy:

Affect Balance = Positive Affect – Negative Affect

An individual who experiences high levels of both positive and negative affect simultaneously (an emotionally volatile life pattern) may maintain a neutral or positive affect balance equal to an individual who experiences low levels of both dimensions (a tranquil or affectively muted life pattern). This construct fundamentally separated emotional intensity and reactivity from emotional balance.

6. Theoretical Framework

The Affect Balance Scale is anchored in the classic Dual-Factor Model of Mental Health and the foundational tenets of modern hedonic psychology. Bradburn’s model emerged as a direct empirical challenge to the classical unitary theory of mood, which held that pleasant and unpleasant emotions occupy opposing ends of a single, bipolar continuum—an assumption dating back to early philosophical treatises and formal psychophysics.

Independence of Affective Systems

The core theoretical axiom of Bradburn’s formulation is the orthogonality (statistical independence) of positive and negative affect. In his large-scale population surveys conducted in four distinct urban and rural communities across the United States, Bradburn made the striking empirical discovery that while items within the positive affect domain correlated robustly with one another, and items within the negative affect domain similarly intercorrelated, the correlation between positive and negative affect subscales was consistently near zero ($r \approx -.05$ to $-.12$).

This empirical independence suggested that the psychological processes that generate joy, achievement, and enthusiasm are neurobiologically and environmentally distinct from the processes that generate anxiety, depression, and social alienation. Consequently, interventions aimed exclusively at reducing psychiatric distress (negative affect) do not automatically generate happiness or fulfillment (positive affect); conversely, fostering positive experiences does not inherently inoculate an individual against depressive episodes.

Biological and Ecological Correspondences

Bradburn’s theoretical insights anticipated modern neuropsychological models of motivation and affect by several decades. Specifically, Bradburn’s dichotomy maps cleanly onto Jeffrey Gray’s Reinforcement Sensitivity Theory (RST):

• Behavioral Approach System (BAS): Governs appetitive motivation, sensitivity to environmental rewards, exploration, and the generation of Positive Affect.
• Behavioral Inhibition System (BIS): Governs sensitivity to threat, punishment, novelty-induced caution, anxiety, and the generation of Negative Affect.

Furthermore, Bradburn demonstrated that Positive Affect is primarily determined by variables reflecting active social life, interpersonal involvement, organizational participation, and novel experiences. In contrast, Negative Affect is predominantly influenced by physical health status, interpersonal friction, financial strain, and perceived socio-structural stressors. The theoretical framework asserts that these two behavioral-ecological systems fluctuate independently across individuals and time.

7. Validity

The construct, criterion, convergent, and discriminant validity of the Affect Balance Scale have been subjected to extensive empirical examination across diverse clinical, non-clinical, geriatric, and cross-cultural cohorts.

Construct and Factorial Validity

Construct validity is heavily anchored in the repeated demonstration of the independence of the two underlying subscales. Across Bradburn’s initial normative samples ($N > 2,000$), correlations between the PA and NA subscales remained non-significant, supporting the postulate that they represent orthogonal constructs rather than poles of a single dimension. Confirmatory structural equation modeling conducted across subsequent decades (e.g., Maitland et al., 2001; Helmes et al., 2010) has consistently supported the two-factor model over unidimensional alternatives.

Convergent Validity

The ABS demonstrates robust convergent validity when compared against external measures of happiness, psychological morale, and psychiatric health:

• Global Self-Reported Happiness: Bradburn demonstrated that net Affect Balance scores correlated strongly ($r = .50$ to $.65$) with single-item direct self-ratings of overall happiness (“Taking all things together, how would you say things are these days?”).
• Geriatric Well-Being and Morale: Moriwaki (1974) confirmed high convergent validity between the ABS and the Philadelphia Geriatric Center Morale Scale ($r = .64, p < .001$) as well as Neugarten’s Life Satisfaction Index-A (LSI-A;$r = .61, p < .001$) in community-dwelling older adults.
• Depression Inventories: The Negative Affect subscale correlates highly ($r = .55$ to $.70$) with standardized depression inventories, such as the Center for Epidemiologic Studies Depression Scale (CES-D) and the Beck Depression Inventory (BDI), whereas the Positive Affect subscale exhibits strong negative associations with clinical anhedonia.

Discriminant Validity

Discriminant validity is confirmed by the distinct socio-environmental correlates of the two subscales. Studies by Beiser, Feldman, and Egelhoff (1972) and subsequent sociological investigations established that:

• Social contact, participation in clubs, visiting friends, and sports participation correlate significantly with Positive Affect, but demonstrate zero correlation with Negative Affect.
• Physical illness symptoms, clinical anxiety, socioeconomic strain, and life crises correlate substantially with Negative Affect, but show virtually no association with Positive Affect.

Predictive and Ecological Validity

Predictive validity has been documented in longitudinal health research. Low baseline Affect Balance scores have been shown to predict elevated risks of functional decline, psychiatric consultation, cognitive impairment, and all-cause cardiovascular mortality in aging cohorts, independent of baseline physical health conditions and biological risk factors.

8. Reliability

The reliability of the Affect Balance Scale has been the subject of substantial psychometric analysis, particularly regarding the trade-offs between internal consistency and temporal stability in brief multidimensional scales.

Internal Consistency Reliability

Estimates of internal consistency for the individual 5-item Positive Affect and Negative Affect subscales are typically modest. Across major psychometric evaluations, Cronbach’s alpha ($lpha$) coefficients generally fall within the following ranges:

• Positive Affect Subscale: $\alpha = .55\text{ to } .73$
• Negative Affect Subscale: $\alpha = .53\text{ to } .71$

While these values fall below the conventional .80 threshold preferred for individual clinical diagnostic testing, psychometric theorists (e.g., McDowell, 2006) note that this reflects the intentional multidimensional breadth of the 5-item indicators rather than measurement error. Bradburn deliberately sampled conceptually diverse facets of affect (e.g., combining psychomotor restlessness, social loneliness, and criticism-induced hurt within a single 5-item index) to capture a broad domain rather than repetitively paraphrasing identical semantic indicators.

Test-Retest Reliability and Temporal Stability

Despite modest internal consistency, the temporal stability of the ABS is robust over short- to medium-term assessment intervals:

• Bradburn (1969) reported test-retest correlation coefficients of $r = .76$ for Positive Affect and $r = .70$ for Negative Affect over a 3-day interval.
• Across a 3-week retest interval, stability coefficients remained moderately high ($r = .65$ to $.70$).
• Over longer multi-month intervals (e.g., 6 to 9 months), stability coefficients decline to approximately $.40\text{ to } .50$, which is theoretically congruent with the nature of affect as a state-like construct that fluctuates in response to changing environmental events and life circumstances.

9. Factor Analysis

The structural dimensionality of the Affect Balance Scale has been rigorously examined through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) across numerous cultural and age-stratified cohorts.

Exploratory Factor Analytic Findings

Early principal component and common factor analyses with varimax (orthogonal) and promax (oblique) rotations conducted by Bradburn (1969), Moriwaki (1974), and Cherlin and Bourque (1981) consistently extracted two primary eigenvalues greater than 1.0, accounting for 40% to 55% of the total variance across items:

• Factor 1 (Positive Affect): Marked by high, unambiguous factor loadings (.50 to .78) from Item 1 (excited/interested), Item 3 (complimented), Item 5 (accomplishment), Item 7 (top of the world), and Item 9 (things going your way). Cross-loadings onto the negative dimension are uniformly trivial ($< .15$).
• Factor 2 (Negative Affect): Marked by substantial factor loadings (.45 to .75) from Item 2 (restless), Item 4 (lonely/remote), Item 6 (bored), Item 8 (depressed/unhappy), and Item 10 (criticized). Cross-loadings onto the positive dimension are consistently negligible ($< .12$).

Confirmatory Factor Analytic (CFA) Evidence

More recent investigations utilizing modern structural equation modeling (SEM) software have formally tested competing measurement models (e.g., Maitland et al., 2001; Helmes, Goffin, & Chrisjohn, 2010):

• Unidimensional Model: A single bipolar factor where positive items load positively and negative items load negatively. This model exhibits extremely poor fit across all conventional criteria (e.g., $\chi^2/df > 6.0$, $text{CFI} < .75$,$text{RMSEA} > .10$).
• Two-Factor Orthogonal Model: Two independent latent variables (Positive Affect and Negative Affect) with covariance fixed to zero. This model yields substantially improved fit across general population samples ($\text{CFI} \approx .91\text{ to } .94$, $\text{RMSEA} \approx .05\text{ to } .06$).
• Two-Factor Oblique Model: Permitting a modest, minor correlation between the two latent factors (typically $r = -.10\text{ to } -.25$) frequently provides the most optimal global fit ($chi^2/df < 2.5$,$text{CFI} > .95$,$text{TLI} > .94$,$text{RMSEA} le .048$,$text{SRMR} le .042$).

These findings conclusively demonstrate that while extreme states of momentary affect may exhibit slight inverse tension, the broad experiential domains of positive and negative well-being represent distinct constructs requiring separate measurement.

10. Instrument / Measurement Tool

• Test Type: Standardized self-report psychological screening questionnaire / rating scale
• Format: Pen-and-paper, face-to-face structured interview, telephone survey, or computerized/online administration
• Number of Items: 10 items (5 Positive Affect items, 5 Negative Affect items)
• Item Configuration:
  • Positive Affect (PA) Subscale: Items 1, 3, 5, 7, and 9
  • Negative Affect (NA) Subscale: Items 2, 4, 6, 8, and 10
• Target Population: General adolescent and adult populations (aged 16+), geriatric cohorts, community-dwelling older adults, family caregivers, and primary care patients
• Completion Time: Approximately 2 to 3 minutes
• Standard Timeframe: “During the past few weeks”
• Response Format: Dichotomous (Yes / No; Yes = 1, No = 0)
• Scoring Procedures:
  • Positive Affect Score (PA): Sum of affirmative (‘Yes’) responses across items 1, 3, 5, 7, and 9. Theoretical range: 0 to 5.
  • Negative Affect Score (NA): Sum of affirmative (‘Yes’) responses across items 2, 4, 6, 8, and 10. Theoretical range: 0 to 5.
  • Affect Balance Score (ABS): Calculated as: $\text{ABS} = \text{PA} – \text{NA}$. Theoretical range: -5 to +5.
  • Transformed Metric (Common Alternative): To eliminate negative numbers in automated computational databases, a constant of 5 is frequently added: $\text{ABS}_{\text{trans}} = (\text{PA} – \text{NA}) + 5$. Theoretical range: 0 to 10.
• Score Interpretation:
  • ABS > 0 (or transformed score > 5): Positive affect balance; reflects overall emotional resilience, subjective well-being, and predominance of positive psychological states.
  • ABS = 0 (or transformed score = 5): Affective equilibrium; indicates equivalent balance between positive and negative affective states.
  • ABS < 0 (or transformed score < 5): Negative affect balance; signals significant emotional distress, potential vulnerability to mood or anxiety disorders, and reduced overall quality of life.

11. Permissions & Fee and Test Year

The Affect Balance Scale was officially published in 1969 by Norman M. Bradburn in his book The Structure of Psychological Well-Being (Chicago: Aldine Publishing Company), following initial pilot exploratory field research published in 1965 with David Caplovitz.

Copyright and Accessibility: As an instrument developed through academic research funded in part by public grants (including the National Institute of Mental Health) and published extensively across public academic domain literature, the Affect Balance Scale is widely considered to be in the public domain for research and clinical assessment purposes. There are no licensing fees, royalty charges, or mandatory commercial purchase requirements associated with administering the 10-item instrument for empirical investigation, educational study, or non-commercial clinical evaluation. Proper academic attribution and citation of Bradburn’s seminal 1969 volume are required in all published reports and formal dissertations incorporating the metric.

12. References

Beiser, M., Feldman, J. J., & Egelhoff, C. J. (1972). Assets and affects: A study of positive mental health. Archives of General Psychiatry, 27(4), 545–549. https://doi.org/10.1001/archpsyc.1972.01750280099017

Bradburn, N. M. (1969). The structure of psychological well-being. Aldine Publishing Company.

Bradburn, N. M., & Caplovitz, D. (1965). Reports on happiness: A pilot study of behavior related to mental health. Aldine Publishing Company.

Cherlin, A., & Bourque, L. B. (1981). Dimensionality and reliability of the Affect Balance Scale in an aggregate survey. Sociological Methods & Research, 9(4), 462–478. https://doi.org/10.1177/004912418100900404

Helmes, E., Goffin, R. D., & Chrisjohn, R. D. (2010). Confirmatory analysis of the Bradburn Affect Balance Scale and its relationship with morale in older Canadian adults. Canadian Journal on Aging / La Revue Canadienne du Vieillissement, 29(2), 259–266. https://doi.org/10.1017/S071498081000017X

Maitland, S. B., Dixon, R. A., Hultsch, D. F., & Hertzog, C. (2001). Well-being as a moving target: Measurement equivalence of the Bradburn Affect Balance Scale. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 56(2), P69–P77. https://doi.org/10.1093/geronb/56.2.P69

McDowell, I. (2006). Measuring health: A guide to rating scales and questionnaires (3rd ed.). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195165678.001.0001

Moriwaki, S. Y. (1974). The Affect Balance Scale: A validity study with aged samples. Journal of Gerontology, 29(1), 73–78. https://doi.org/10.1093/geronj/29.1.73

Perkinson, M. A., Albert, S. M., Luborsky, M., Moss, M., & Glicksman, A. (1994). Exploring the validity of the Affect Balance Scale with a sample of family caregivers. Journal of Gerontology, 49(5), S264–S275. https://doi.org/10.1093/geronj/49.5.S264

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063

13. Items of the Scale (Questionnaire)