Research figure: Albert Bandura · Written by Artificial Sapiens
Why the 1977 paper matters
Albert Bandura's 1977 article “Self-Efficacy: Toward a Unifying Theory of Behavioral Change” is a foundational statement of how beliefs about one's ability to organize and execute a particular action may shape the initiation, effort, persistence, and course of behavior. Published in Psychological Review, volume 84, issue 2, pages 191–215, the article introduced an explicitly task-sensitive account of perceived self-efficacy and proposed it as a common psychological mechanism through which apparently different therapeutic procedures can produce behavioral change. Its author was Albert Bandura alone. The paper's historical significance lies in joining observations about successful action, social modeling, persuasion, and emotional arousal to a testable theory of perceived capability (Bandura, 1977; National Library of Medicine, 1977).
The article's central question was consequential: why do people who know which actions are possible, and sometimes know which actions would produce a desirable result, nonetheless fail to initiate them or abandon them when difficulties appear? Bandura argued that an account of behavioral change must include a person's judgment that they can perform the relevant action under the actual circumstances. His theory also made a further empirical proposal: experiences that change these judgments should help explain subsequent changes in behavior. Whether efficacy beliefs themselves cause improved performance, rather than merely record previous success or other changes, remains a question that requires careful research design (Bandura, 1977; Sitzmann and Yeo, 2013).
This article examines the 1977 paper as an original scientific work: its precise thesis, intellectual context, argument, sources of efficacy information, use of fear and phobia research, measurement logic, influence, and what later evidence does and does not establish. A broader introduction to the researcher is available in Albert Bandura, while the general construct belongs to the separate article Self-Efficacy: How Belief in Your Ability Shapes Action.
The original article: bibliographic identity and scientific genre
The exact original title is “Self-Efficacy: Toward a Unifying Theory of Behavioral Change,” published in March 1977 by the American Psychological Association in Psychological Review. The bibliographic record is 84(2), 191–215, DOI 10.1037/0033-295X.84.2.191, PMID 847061. The work is a theoretical integration supported by discussion and microanalysis of behavioral-treatment research. It is therefore best read as a proposed explanatory framework incorporating empirical observations, rather than as one stand-alone randomized clinical trial establishing a universal mechanism (PubMed bibliographic record; ERIC record).
The phrase “unifying theory” identified an explanatory ambition. Behavioral therapies, exposure-like procedures, participant modeling, vicarious observation, persuasive interventions, and methods directed at emotional arousal were often discussed in different therapeutic vocabularies. Bandura proposed that their consequences might be compared by asking whether each procedure changes perceived personal capability. He did not demonstrate that every successful psychological intervention works exclusively through self-efficacy. That stronger proposition would require independently measured mediators and designs capable of distinguishing efficacy change from habituation, new skill acquisition, altered danger appraisal, therapist contact, and practice (Bandura, 1977).
The paper appeared in the same year as Bandura's book Social Learning Theory, but the two are separate works. The article's distinctive focus is the efficacy mechanism of behavioral change and its relation to treatment studies; the book had a broader learning-theory agenda. Another crucial distinction concerns the empirical companion paper “Cognitive Processes Mediating Behavioral Change,” authored by Albert Bandura, Nancy E. Adams, and Janice Beyer, published in Journal of Personality and Social Psychology, 35(3), 125–139. Claims about that experiment's specific treatment arms and observed outcomes must be attributed to all three authors, rather than presented as findings of a solo-authored experiment (Bandura, Adams, and Beyer, 1977).
The problem Bandura set out to explain
Knowledge of consequences versus perceived ability
A person may expect that approaching a feared animal, speaking to a group, completing a demanding assignment, or following a health behavior will yield a beneficial consequence, while doubting that they can complete the necessary actions. Bandura distinguished outcome expectations—judgments that a behavior will produce particular consequences—from efficacy expectations—judgments about one's own capability to perform the behavior. The two evaluations may be related, but they concern different propositions. A person might believe that a treatment technique works in general and still doubt their ability to carry it out. Conversely, a person might feel able to perform an action yet see little value in its likely outcome (Bandura, 1977).
This distinction clarified why information and incentives can be insufficient for action. Expected reward cannot bring about an act that the person believes is beyond their capacity; perceived capacity can influence which opportunities are attempted in the first place. The paper thus located a consequential appraisal between environmental opportunities, learned behavioral repertoires, and action. The term “belief” here describes a judgment with a concrete behavioral referent, rather than a detached feeling of optimism.
Efficacy expectations and behavior
Bandura proposed that perceived self-efficacy affects whether a person initiates a course of action, how much effort they invest, and how long they persevere when demands become difficult. In threatening situations, a low efficacy expectation may encourage anticipatory fear or avoidance; stronger efficacy expectations may support sustained coping attempts. The model presumes that the relevant action is feasible and that environmental constraints permit its execution. Confidence cannot independently create unavailable tools, acquired expertise, physical capacity, institutional access, or safety (Bandura, 1977).
Crucially, the claim is probabilistic and task-dependent. The original conception does not authorize interpreting high self-efficacy as proof that someone has the required competence, and it does not imply that a person should persist indefinitely in dangerous or fruitless circumstances. Accurate appraisal of actual task demands, available supports, and changing feedback matters to the quality of decisions.
What self-efficacy meant in the original formulation
Self-efficacy in this paper refers to expectations of personal capability for specified performances. Bandura used efficacy judgments to connect a person's perceived repertoire to anticipated difficulty and action. A good reading retains three levels of analysis: the particular behavior, the conditions under which it is attempted, and the strength or certainty of the perceived ability. The relevant question might be whether a person can approach an animal from a specified distance, touch it with support, or remain near it unaided. Those tasks vary in demand; success at one level need not imply success at all levels (Bandura, 1977).
Magnitude, strength, and generality
The 1977 framework described efficacy expectations by their magnitude, strength, and generality. Magnitude concerns the difficulty of tasks a person believes they can perform. Strength concerns how firmly the person holds that judgment, including resilience when an initial attempt does not succeed. Generality concerns whether an efficacy judgment extends across tasks or conditions, including situations sharing meaningful demands. These dimensions make a global yes-or-no question such as “Are you confident?” a poor substitute for a careful task-specific assessment.
The dimensions are conceptually distinct. Someone might rate their capability highly for easy tasks but weakly for difficult tasks. Someone else's judgment might be robust across several threatening situations while remaining narrow to a single type of activity. Researchers can represent these distinctions with graded tasks and certainty ratings; however, the quality of any measure depends on its relevance to the observed behavior and on how the tasks are sampled.
Why perceived capability differs from ability
Perceived efficacy is a psychological variable; demonstrated ability is a performance characteristic. The judgment can be calibrated well, overly optimistic, or overly pessimistic. Practice may improve skills and efficacy beliefs together, creating a statistical association without proving which change caused the other. Likewise, a new technique can eliminate a barrier to performance even when belief change is limited. The original article proposed efficacy as a major mediator; it did not dissolve the conceptual distinction between skill, resources, perceptions, and behavior. Modern studies of reciprocal efficacy–performance relations make this distinction especially important (Sitzmann and Yeo, 2013; Talsma et al., 2018).
Four major sources of efficacy information
Bandura's 1977 account identified four broad sources of information from which people construct efficacy expectations. These sources are causal hypotheses about information processing, not four mutually exclusive treatment brands. Their relevance depends on the task, the credibility of the information, the observer's prior experience, and interpretation of what happened (Bandura, 1977).
Performance accomplishments and enactive mastery
Experiencing successful performance provides direct evidence that an action is possible for the person who performed it. Repeated successes under increasingly demanding but manageable conditions can build a stronger expectation of future capability. Failures may weaken the expectation, particularly early in a learning sequence, although their meaning depends on difficulty, effort, external circumstances, and a person's interpretation of the attempt. Mastery is especially informative when the observed performance resembles the target behavior and the person recognizes their own contribution to success.
In the phobia research informing the paper, treatment procedures sometimes helped participants approach a feared object in small stages with guidance. Such action could both furnish efficacy information and produce behavioral learning, altered threat expectations, and habituation. Because these processes can unfold together, improvement after successful practice supports the practical significance of mastery-based experiences more directly than it identifies a single exclusive mediator (Bandura, Adams, and Beyer, 1977).
Vicarious experience
Seeing another person perform a feared or demanding action can influence one's estimate of what is attainable. Similarity of the model, perceived relevance of the demonstration, observed consequences, and whether the observer can reproduce the required steps all affect its informational value. Observation may suggest possible methods or reduce anticipated difficulty without giving the observer the same evidence as successful first-person performance. In a therapy setting, the ability to watch safe handling of a feared stimulus can be useful even before direct practice begins (Bandura, 1977).
The distinction between observing a successful performance and performing it is essential. Vicarious evidence can strengthen willingness to attempt a task, while mastery can also change actual skill and provide feedback about one's own capacities. Their effects should be compared with respect to the task and the intervention rather than assumed to be identical.
Verbal persuasion
Encouragement, explanations, instruction, and feedback can make a capability judgment more favorable when the information appears credible. Bandura's account also identified a limitation: persuasion unsupported by real opportunities to act has a weaker foundation than demonstrated mastery and may be undermined by subsequent failure. Contemporary applications sometimes mistake this source for repeated positive affirmations. The 1977 framework instead directs attention to credibility, realistic task demands, and behavioral corroboration (Bandura, 1977).
Persuasion can be consequential when it communicates concrete strategies, identifies a manageable next step, or corrects a mistaken interpretation of prior experience. In those situations, its effect is entwined with instructional content and actual task conditions. An experimental evaluation needs to separate supportive words from the skills and opportunities that accompany them.
Physiological states and emotional arousal
People may interpret bodily arousal, tension, fatigue, and emotional disturbance as evidence that a task is threatening or beyond their ability. The 1977 formulation discussed emotional and physiological information as one channel through which efficacy expectations are formed. A reduction in fear-associated arousal may therefore change perceived capability, but the same arousal can also reflect actual danger, exertion, illness, or normal task engagement. Its meaning depends on situation and interpretation (Bandura, 1977).
This part of the theory is best read historically. The 1977 text used the language of physiological states and emotional arousal; later teaching often develops the category under revised terminology. The important scientific claim remains that people use bodily and affective evidence to judge capacity, with its usefulness conditional on the source and interpretation of those signals.
The article's central behavioral-change model
The paper connected intervention experience to perceived efficacy and then to coping action. Its theoretical sequence can be expressed as a chain: a person encounters performance-related information; the information is interpreted in relation to an attainable task; an efficacy expectation forms or changes; the expectation contributes to whether the person attempts and sustains the action; the resulting performance supplies additional information. This is a dynamic account because feedback from successful or unsuccessful behavior can change later beliefs (Bandura, 1977).
The model is valuable precisely because it creates empirical questions. Do two treatments producing comparable skill gains but different efficacy judgments yield different behavior? Does a change in efficacy precede a subsequent performance change when earlier performance is controlled? Can a manipulated efficacy message alter behavior without changing opportunities or skills? When a person succeeds, does the experience raise efficacy because the person correctly infers increased capacity, because the challenge was easier than expected, or because another aspect of the context changed? Different answers imply different mechanisms.
A mediating variable must be evaluated with more than its correlation with the outcome. An efficacy rating recorded just before a behavior may predict that behavior while still reflecting earlier treatment effects, knowledge, avoidance tendencies, and prior successes. Stronger causal tests require temporal ordering, appropriate comparison conditions, direct measurement of competing processes, and, where ethical and feasible, experimental manipulation of components designed to change perceived efficacy independently of skill.
Fear and phobia research as the original empirical setting
The clinical illustrations are central to interpreting the paper. Work with fear of snakes and related approach tasks created situations in which perceived capability could be measured as a graded expectation and behavior observed as a graded approach. This close match between belief item and behavioral test offered a methodological advantage over asking people whether they were generally self-confident. A participant could report whether they expected to approach or handle an object at successive levels of difficulty, and the researchers could then observe which activities the participant completed (Bandura, 1977).
Guided mastery, modeling, and behavioral approach
Participant modeling and graduated performance tasks combined social demonstration, assistance, successive challenges, and individual action. These are clinically meaningful procedures, but they change several ingredients at once. Watching a model can reduce uncertainty about what will happen. A therapist's help may reduce immediate task difficulty. A completed action can provide a corrective experience. Reduced avoidance can permit further practice. Improved performance can consequently be associated with improved efficacy for multiple reasons.
The 1977 evidence supports the conclusion that efficacy ratings covaried closely with tested coping performance in the studied conditions and were sensitive to different methods of treatment. It does not, by itself, establish identical mechanisms across all anxiety disorders, therapy formats, and populations. The historical laboratory procedures must also be distinguished from current disorder-specific clinical guidelines and full-course psychotherapy outcomes.
The distinct 1977 experiment by Bandura, Adams, and Beyer
A particularly important contemporaneous study is “Cognitive Processes Mediating Behavioral Change.” Bandura, Adams, and Beyer compared conditions involving performance mastery, vicarious experience, and no treatment among adults with phobic fears. Their outcomes included efficacy expectations and approach behavior toward threats with varying degrees of similarity to treatment targets. The authors reported that mastery-based treatment produced stronger and more generalized efficacy judgments than vicarious experience alone, and that graded efficacy ratings closely predicted subsequent performance within their studied tasks (Bandura, Adams, and Beyer, 1977; PubMed record).
This experiment lends empirical weight to the theoretical paper, while carrying its own methods, sample, treatment arms, and limitations. It is inaccurate to credit its three-author experimental results solely to the single-author Psychological Review article. It is equally inaccurate to infer from successful prediction within graded phobia tasks that the same coefficient or causal ordering necessarily applies to academic achievement, health behavior, organizational performance, or long-term relapse prevention.
The measurement logic introduced by the paper
Bandura's approach made perceived capability assessable through questions aligned with observable actions. An item such as “Can you take a specified step under these conditions?” has greater construct relevance than a global confidence statement if the research outcome is performance of that exact step. Graded task levels allow researchers to record where confidence begins to fall and how firmly a person endorses each expected level of success. Such measures produce interpretable predictions only when the difficulty scale and situations actually correspond to the behavior being studied.
Behavioral specificity and correspondence
A self-efficacy instrument for approaching feared objects is not automatically valid as a measure of writing essays or maintaining exercise. Task descriptions, response certainty, and temporal context must be adapted to the target. The principle of correspondence later became important across education and sport: self-efficacy is often more strongly associated with performance when predictor and outcome refer to the same domain and comparable demands. In a meta-analysis of sport performance, Moritz and colleagues found an average correlation of .38 across 45 studies and identified alignment between efficacy and performance measures as a major moderator (Moritz et al., 2000).
A close match improves relevance, but it can also create conceptual and methodological problems. If efficacy questions simply ask respondents to forecast the imminent outcome in terms nearly identical to the performance test, predictive accuracy may partly reflect their accurate knowledge of current capacity. Researchers must distinguish predictive correspondence from explanatory independence. A high correlation alone does not determine whether beliefs caused the later behavior.
Temporal order and shared-method risks
Measurement before action is useful but not sufficient for causality. Prior performance can influence the rating, and the same unmeasured capacity or contextual feature can influence both. Self-reported efficacy and self-reported success are additionally susceptible to shared response tendencies. Strong studies use behavioral or independently verified outcomes when possible, consider baseline competence and task difficulty, and investigate whether belief change precedes performance change across repeated observations.
What the 1977 argument did—and did not—establish
The 1977 paper established a historically influential conceptual distinction and a program of operationalizable propositions. It gathered therapeutic phenomena into a coherent account and discussed empirical microanalyses consistent with that account. The direct record is particularly strong for showing that graded efficacy judgments can accompany and predict graded coping performance in the specific situations studied. This is more precise than saying the paper proved a universal psychological law (Bandura, 1977).
The paper did not establish that all methods of psychotherapy act through the same mediator or that confidence is the only determinant of behavioral change. It did not provide a population-level estimate of self-efficacy's causal influence across domains. It did not demonstrate that higher belief is always beneficial, that every person interprets similar mastery information identically, or that short-term improvement in behavioral approach necessarily yields durable remission from a clinical disorder. Those would be distinct empirical claims requiring separate designs and follow-up.
An interpretive caution concerns the word “predict.” In the original clinical microanalyses it refers to correspondence between measured efficacy expectations and specific performances. In causal inference, prediction, mediation, and intervention effects are different questions. The historical material warrants close attention to this distinction rather than the assumption that predictive strength resolves mechanism.
How later research tested the central prediction
Within-person meta-analysis: Sitzmann and Yeo
Sitzmann and Yeo (2013) examined the within-person relationship between self-efficacy and performance, comparing belief-to-performance with prior-performance-to-belief associations. They reported a corrected within-person efficacy–performance correlation of .23, which decreased to .06 and was not statistically significant after controlling for a linear trajectory. The corresponding prior-performance-to-efficacy association remained positive after the same adjustment (ρ = .30). This finding challenges simple interpretations in which an observed increase in efficacy independently drives all subsequent improvement. Its inference is bounded by the included tasks, repeated-measure designs, modeling assumptions, and controls; it does not erase evidence from other experimental or domain-specific designs (Sitzmann and Yeo, 2013).
The methodological point is substantial. People often improve through practice over time, and their efficacy beliefs improve at the same time. Failure to model common time trends can make a forward belief–performance association appear more autonomous than it is. The result gives investigators a reason to measure performance history and time-varying processes explicitly.
Longitudinal educational evidence: Talsma and colleagues
Talsma, Schüz, Schwarzer, and Norris (2018) combined longitudinal studies of academic self-efficacy and achievement in a meta-analytic cross-lagged panel analysis. The analysis included 11 studies and 2,688 participants. Prior performance predicted subsequent efficacy more strongly (β = .205) than efficacy predicted subsequent performance (β = .071), although both estimates were statistically distinguishable from zero in the aggregate model. Reciprocity appeared in adult samples, with a different pattern in children. The results support domain- and population-sensitive reciprocal relationships, while observational panel models by themselves cannot eliminate every unmeasured time-varying confounder (Talsma et al., 2018).
The apparent tension between this finding and the Sitzmann–Yeo analysis is scientifically informative. The samples, performance domains, time structures, adjustments, and statistical models differ. Neither should be presented as the final universal test of Bandura's entire program. Taken together, they show why the direction and magnitude of effects have to be examined rather than presumed.
Evidence from sport performance
Moritz, Feltz, Fahrbach, and Mack (2000) synthesized 45 sport studies comprising 102 correlations and found an average efficacy–performance association of .38. The relationship varied across tasks and measures, and close alignment between efficacy assessments and outcomes mattered. This is evidence for meaningful association within sport, not proof of a uniform causal coefficient. Athletes may infer their efficacy from prior skill and training, and higher efficacy can sometimes support selection of challenging practice tasks; well-designed experiments are needed to separate these processes (Moritz et al., 2000).
Later expansion by Bandura
Bandura's later work enlarged the scope from therapeutic behavior to self-regulation, collective action, social cognition, and health promotion. His 2001 “Social Cognitive Theory: An Agentic Perspective” elaborated human agency in a broader theoretical architecture, while his 2004 “Health Promotion by Social Cognitive Means” extended social cognitive principles into health behavior. These publications are later developments and applications, not sections retroactively belonging to the 1977 article (Bandura, 2001; Bandura, 2004).
Scientific appraisal by evidence level
The established bibliographic facts are the article's authorship, journal, date, pages, and central theoretical propositions. Its distinction between perceived capacity and expected consequences is a precise and influential conceptual contribution. Evidence is also well established that self-efficacy measures are associated with performance in multiple studied domains, although the association varies with task and method (Bandura, 1977; Moritz et al., 2000).
Evidence is context-dependent concerning the causal contribution of belief change over and above skill, experience, opportunity, and common performance trends. The graded phobia studies support a meaningful relationship under studied conditions; longitudinal and within-person syntheses show that directionality and adjustment matter. The broad assertion that efficacy is the sole or universal mediator of psychotherapy is not established by the 1977 paper or by the later reviews considered here (Bandura, Adams, and Beyer, 1977; Sitzmann and Yeo, 2013; Talsma et al., 2018).
The paper's wider historical influence on intervention design and psychological theorizing is established through its central role in subsequent scholarship. Claims that any particular present-day intervention works for a diagnosed condition require evidence on that intervention and condition, including patient-important outcomes, harms, and follow-up. A mechanistic theory cannot substitute for those studies.
Historical significance and intellectual influence
A bridge between behavioral procedures and cognitive explanation
The paper addressed a period of intensive inquiry into how psychological treatment works. Behavioral procedures had demonstrated that contact with feared situations, guided rehearsal, and modeled action could alter behavior. Bandura supplied an account of the information people extract from these experiences and the ensuing expectations about personal capability. His argument encouraged investigators to measure the participant's interpretation of treatment, not only the outward exposure or final behavioral outcome (Bandura, 1977).
The account was especially consequential because it articulated why the same external procedure might affect participants differently. Success, persuasion, observation, and reduced arousal are informative only through their interpretation in relation to the person's target activity. This hypothesis opened questions about feedback, appraisal, graded challenge, and transfer across tasks. At the same time, a contemporary researcher can study these mechanisms without assuming that one theoretical vocabulary captures every effective component of an intervention.
From clinical coping to education and performance
Researchers adapted task-specific efficacy judgments to classroom tasks, skill development, sport, health behavior, organizational learning, and other domains. Every transfer raises a construct-validity problem: what counts as the relevant action, how task difficulty is represented, and whether a new performance measure actually corresponds to the belief scale. The original phobia setting offered unusually concrete graded tasks; other settings frequently contain shifting goals and shared institutional constraints. Applications are therefore scientific extensions requiring their own evidence rather than replications of the original phobia findings (Moritz et al., 2000; Talsma et al., 2018).
For education specifically, the encyclopedia's Academic Self-Efficacy: Confidence, Effort, and Achievement covers that applied domain. The 1977 work-specific article instead preserves the theoretical source and its evidentiary logic.
Relation to later social cognitive theory
The 1977 article anticipates subsequent developments in social cognitive thinking, but the expanded mature framework should be dated and attributed accurately. Bandura later articulated a more extensive architecture of personal factors, behavior, and environmental influences, together with accounts of self-regulation and agency. A reader seeking the general theoretical comparison across traditions can consult Social-Cognitive Personality Theory: Reciprocal Determinism and Behavior. This division preserves the 1977 paper as a discrete primary work rather than converting it into a substitute for all of Bandura's later theory (Bandura, 2001).
Frequently asked questions about the 1977 paper
What was the exact hypothesis of Bandura's 1977 self-efficacy paper?
Bandura proposed that people's judgments about their capability to perform specific actions contribute to whether they undertake those actions, the effort they invest, and their persistence under challenge. He further proposed that different behavioral-change procedures may work in part by altering those efficacy judgments. The latter is a mediation hypothesis, requiring evidence about the mechanism rather than a simple association (Bandura, 1977).
Was the article an experiment or a theoretical review?
The Psychological Review article was a theoretical integration informed by empirical examples and microanalyses. It should be distinguished from the separately published experimental study conducted by Bandura, Adams, and Beyer in Journal of Personality and Social Psychology in 1977. The two papers are linked intellectually but have different author lists and scientific functions (PubMed: original paper; PubMed: three-author experiment).
Which four sources of efficacy information did Bandura identify?
He discussed performance accomplishments, vicarious experience, verbal persuasion, and physiological states or emotional arousal. Each source provides information that may change a task-specific judgment of capability, with actual mastery generally offering more direct personal evidence than observation or verbal assurance (Bandura, 1977).
Is self-efficacy the same as self-esteem, confidence, or motivation?
The core construct is a belief about the ability to execute specified actions in specified conditions. Self-esteem concerns self-evaluation; motivation is a broader family of processes influencing action; confidence without a task referent is less precise. Self-efficacy may influence motivation, and motivation may affect training and feedback, but the terms do not identify the same variable.
Did Bandura prove that believing in yourself always improves performance?
The 1977 paper proposed a mechanism and discussed supportive findings in particular treatment contexts. Later quantitative research documents varied associations and reciprocal relations between prior performance and subsequent efficacy, with some adjusted effects becoming weak. Claims of universal benefit or exclusive causation go beyond the evidence (Sitzmann and Yeo, 2013; Talsma et al., 2018).
Why were phobias especially useful for the original research?
Graded approach tasks permitted researchers to compare specific efficacy expectations with subsequent behavior across different levels of difficulty. This task correspondence made prediction more interpretable and connected theoretical judgments to observable acts. Yet efficacy ratings, actual skill learning, and reduced fear could change together during treatment, so causal claims require more than successful prediction (Bandura, Adams, and Beyer, 1977).
What is the most accurate way to cite the paper?
Cite Albert Bandura as sole author, the year 1977, the exact title “Self-Efficacy: Toward a Unifying Theory of Behavioral Change,” Psychological Review, 84(2), 191–215, and DOI 10.1037/0033-295X.84.2.191. Cite Bandura, Adams, and Beyer separately when referring to their three-author phobia experiment (National Library of Medicine, 1977).
Conclusion
Bandura's 1977 self-efficacy paper transformed the scientific study of behavioral change by proposing that perceived capability for specific actions should be measured and incorporated into explanations of effort, avoidance, persistence, and therapeutic improvement. Its four sources of efficacy information and its distinction between efficacy and outcome expectations remain central to the historical definition of the construct. The most durable scientific reading combines that conceptual achievement with precise limits: graded efficacy judgments can inform prediction; practice and performance can alter the judgments themselves; and claims about independent causal mediation require stronger tests than prediction alone. The paper is consequently both an influential original theoretical work and an enduring invitation to investigate the dynamic relationship between beliefs, capabilities, environments, and actions.
Related Articles
Albert Bandura — A comprehensive scientific account of Bandura's career, original experiments, theories, and intellectual legacy.
Self-Efficacy: How Belief in Your Ability Shapes Action — The general psychological construct, its practical meaning, and relevant evidence across contexts.
Academic Self-Efficacy: Confidence, Effort, and Achievement — Domain-specific research concerning efficacy beliefs and academic outcomes.
Social-Cognitive Personality Theory: Reciprocal Determinism and Behavior — The wider conceptual framework and its relationship to personality research.
Albert Bandura's Self-Efficacy: The Exercise of Control — The comprehensive scientific monograph published in 1997.
Albert Bandura's Self-Efficacy Theory — Capability beliefs, efficacy information, and theoretical mechanisms.
References
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1–26. https://doi.org/10.1146/annurev.psych.52.1.1
Bandura, A. (2004). Health promotion by social cognitive means. Health Education and Behavior, 31(2), 143–164. https://pubmed.ncbi.nlm.nih.gov/15090118/
Bandura, A., Adams, N. E., and Beyer, J. (1977). Cognitive processes mediating behavioral change. Journal of Personality and Social Psychology, 35(3), 125–139. https://doi.org/10.1037/0022-3514.35.3.125
Moritz, S. E., Feltz, D. L., Fahrbach, K. R., and Mack, D. E. (2000). The relation of self-efficacy measures to sport performance: A meta-analytic review. Research Quarterly for Exercise and Sport, 71(3), 280–294. https://doi.org/10.1080/02701367.2000.10608908
Sitzmann, T., and Yeo, G. (2013). A meta-analytic investigation of the within-person self-efficacy domain: Is self-efficacy a product of past performance or a driver of future performance? Personnel Psychology, 66(3), 531–568. https://doi.org/10.1111/peps.12035
Talsma, K., Schüz, B., Schwarzer, R., and Norris, K. (2018). I believe, therefore I achieve (and vice versa): A meta-analytic cross-lagged panel analysis of self-efficacy and academic performance. Learning and Individual Differences, 61, 136–150. https://doi.org/10.1016/j.lindif.2017.11.015