Responsibility for Student Achievement (RSA)

1. Abstract

The Responsibility for Student Achievement (RSA) questionnaire is a seminal psychometric instrument developed by Thomas R. Guskey (1981, 1988) designed to measure teachers’ locus of control and causal attributions regarding student academic outcomes. Operating at the intersection of attribution theory and teacher efficacy, the RSA assesses the degree to which educators assume personal responsibility for classroom successes versus classroom failures. The scale consists of 30 forced-choice dyadic items utilizing a constant-sum, percentage-allocation response format where respondents distribute 100 percentage points between an internal attribution (teacher instructional practices, planning, motivation, or effort) and an external attribution (student ability, inherent motivation, peer influence, or parental background).

The instrument yields three primary psychometric scores: responsibility for student success (R+), responsibility for student failure (R-), and overall responsibility (Total R). Extensive empirical investigations across elementary, middle, and secondary education settings demonstrate that teachers’ attributional patterns are multidimensional and asymmetrical; educators typically demonstrate a self-serving attributional bias by internalizing positive outcomes (R+) while attributing academic difficulties or instructional breakdown (R-) to external sources. Psychometric evaluations confirm satisfactory internal consistency, with Cronbach’s alpha coefficients typically ranging from .72 to .85 across subscales, and moderate test-retest reliability across multi-month intervals. Construct, convergent, and discriminant validity analyses substantiate the RSA as an essential diagnostic and evaluative tool in teacher education, educational reform, professional development efficacy research, and school psychology.

2. Keywords

Responsibility for Student Achievement, RSA, Thomas R. Guskey, teacher efficacy, causal attributions, locus of control, academic achievement, mastery learning, constant-sum scale, teacher responsibility, educational psychometrics, instructional psychology.

3. Authors

The Responsibility for Student Achievement (RSA) questionnaire was conceptualized, operationalized, and validated by Dr. Thomas R. Guskey. Dr. Guskey is Professor Emeritus of Educational Psychology in the College of Education at the University of Kentucky. A prominent scholar in educational measurement, grading systems, professional development evaluation, and mastery learning, Guskey earned his Ph.D. in Measurement, Evaluation, and Statistical Analysis from the University of Chicago under the mentorship of Benjamin S. Bloom.

Throughout his academic career, Dr. Guskey served in major leadership capacities within the American Educational Research Association (AERA) and the National Council on Measurement in Education (NCME). His seminal 1981 validation paper in the Journal of Teacher Education, alongside subsequent refinements in 1982 and 1988, established the RSA as a critical instrument in teacher cognition and school change dynamics. Correspondence regarding the scale’s historical development and theoretical architecture is indexed via university repositories and academic educational measurement archives.

4. Purpose

The core objective of the Responsibility for Student Achievement (RSA) questionnaire is to measure the internal versus external attributional orientations of practicing and pre-service educators with respect to pedagogical successes and failures. Historically, educational researchers observed that systemic professional development programs often failed to produce enduring changes in teacher classroom behaviors. Guskey hypothesized that this resistance was anchored in teacher beliefs regarding locus of responsibility: teachers who perceive student learning outcomes as predominantly governed by immutable student traits (such as socioeconomic background, innate intellectual ability, or home environment) are fundamentally less likely to adopt innovative instructional strategies or engage in rigorous pedagogical self-reflection.

The RSA addresses several critical measurement, research, and applied clinical purposes:

• Evaluating Differential Attributions for Success vs. Failure: Many generic locus of control instruments (such as Julian Rotter’s Internal-External scale) treat perceived control as a unidimensional, bipolar continuum. The RSA purposely bifurcates teacher attributions into distinct success events (R+) and failure events (R-). This operational differentiation exposes whether an educator accepts agency for classroom mastery while systematically disowning pedagogical culpability for student failure.
• Predicting Teacher Receptivity to Instructional Innovation: Research utilizes the RSA to predict how educators respond to mastery learning interventions, differentiated instruction, and structured feedback systems. Educators with elevated R+ and R- scores consistently display significantly greater readiness to modify their teaching practices, engage in data-driven decision-making, and persist when instructional tasks prove challenging.
• Assessment of Professional Development Interventions: Guskey’s linear model of teacher change indicates that substantial shifts in teacher attitudes and attributional beliefs occur primarily after educators witness observable improvements in student learning outcomes resulting from modified pedagogical practices. The RSA serves as a sensitive pre- and post-test measurement device to track these longitudinal cognitive shifts.
• Diagnostic Tool for Educator Burnout and Attrition: Teachers who exhibit low perceived responsibility combined with pervasive external blame often report diminished occupational satisfaction, emotional exhaustion, and feelings of learned helplessness within high-stakes accountability environments. Conversely, maladaptive hyper-internalization of failure without institutional support can precipitate acute performance anxiety and professional fatigue.

5. Psychological Construct

The psychological construct evaluated by the RSA is Teacher Attributional Responsibility for Academic Outcomes. This construct operates at the nexus of locus of control, social cognitive agency, and cognitive appraisal theory. Unlike student academic attribution scales (e.g., the Intellectual Achievement Responsibility questionnaire), the RSA situates the adult educator as the evaluating agent determining the proximal causation of classroom phenomena.

Responsibility for Success (R+)

The R+ subscale assesses the extent to which a teacher attributes positive classroom events, student mastery, cognitive breakthroughs, and institutional commendations to their own instructional efforts, pedagogical clarity, classroom management, and affective encouragement. A teacher scoring high on R+ interprets a student scoring highly on an exam not as an artifact of an excessively simplistic evaluation or innate intellectual giftedness, but as a direct consequence of well-organized content, transparent performance expectations, and intentional instructional scaffolding.

Illustrative manifestations of R+ attribution include:

• Attributing student enthusiasm and engagement to dynamic lesson design and collaborative learning structures rather than situational novelty.
• Ascribing a struggling student’s sudden conceptual clarity to systematic formative assessment, diagnostic feedback loops, and persistent re-explanation.
• Interpreting administrative praise as validation of sustained classroom instructional efficacy rather than arbitrary administrative politeness.

Responsibility for Failure (R-)

The R- subscale quantifies the degree to which an educator accepts personal responsibility when students fail to comprehend concepts, perform poorly on standardized or classroom-based assessments, disengage from academic tasks, or exhibit behavioral resistance. Rather than dismissing failure as the natural outcome of cognitive deficits, socioeconomic challenges, or motivational apathy, a high R- educator internalizes student difficulty as an indicator that the instructional method, pacing, diagnostic assessment, or prerequisite reinforcement was insufficient.

Representative operationalizations of R- attribution encompass:

• Viewing widespread student failure on an examination as evidence of inadequate instructional pacing, unclear alignment between objectives and assessment, or poor preparation.
• Explaining student inattentiveness as a symptom of lesson pacing or presentation design, rather than chronic student deficit.
• Interpreting parental criticism as reflecting a failure in reciprocal communication and instructional differentiation rather than parental hostility.

Total Responsibility (Total R)

The composite Total R index reflects an educator’s cumulative internal locus of control across the entire educational continuum. Teachers with elevated Total R scores manifest what cognitive psychologists characterize as an active master-oriented instructional posture: they believe that pedagogical decisions are the primary determinant of student learning trajectories, transcending demographic or structural barriers.

6. Theoretical Framework

The Responsibility for Student Achievement questionnaire is anchored in three primary theoretical traditions: Bernard Weiner‘s Attribution Theory of Motivation, Julian Rotter‘s Social Learning Theory, and Albert Bandura‘s Self-Efficacy Theory.

Weiner’s Causal Attribution Model

Weiner’s taxonomy categorizes causal attributions along three orthogonal dimensions: locus of causality (internal vs. external), stability (stable vs. unstable over time), and controllability (controllable vs. uncontrollable by the actor). Guskey adapted this architecture directly to classroom dynamics. The RSA focuses heavily on the locus of causality and controllability dimensions. When an educator attributes student success to teacher effort or pedagogical design, the cause is internal and controllable. Conversely, when success is attributed to student innate talent, the cause is external to the teacher. In failure conditions, attributing student difficulties to the teacher’s insufficient planning involves an internal, unstable, and controllable attribution, which theoretically promotes future corrective behavioral change.

Rotter’s Locus of Control & the IAR Precedents

Rotter’s conceptualization of generalized expectancies for internal versus external reinforcement established that individuals differ systematically in whether they view life outcomes as contingent upon their own actions or external forces. Crandall, Katkovsky, and Crandall (1965) advanced this framework by developing the Intellectual Achievement Responsibility (IAR) questionnaire for children, demonstrating that internal attributions for success and failure do not correlate perfectly and must be scored independently. Guskey recognized that adult instructional psychology mirrored this finding: teachers do not possess a monolithic internal-external locus; rather, their sense of agency fluctuates systematically based on outcome valence.

Bandura’s Social Cognitive and Efficacy Formulations

Although developed concurrently with the earliest measures of teacher self-efficacy (such as the RAND Corporation’s two-item scale and subsequent work by Ashton and Webb), the RSA captures an indispensable cognitive antecedent to efficacy expectations. As Bandura articulated, outcome expectations—beliefs that specific behaviors will produce designated outcomes—are inextricably linked to causal attributions. If a teacher maintains that external variables (heredity, peer norms) overpower pedagogical influence, high personal teaching efficacy cannot functionally develop. The RSA provides the empirical measurement of this foundational attributional terrain.

7. Validity

The psychometric validity of the RSA has been corroborated through rigorous construct, convergent, criterion-related, and discriminant validity investigations across multiple decades of educational inquiry.

Construct and Structural Validity

Construct validation studies initially conducted by Guskey (1981, 1982) demonstrated that R+ and R- function as quasi-independent attributional dimensions rather than polar opposites on a single continuum. The correlation between R+ and R- subscales typically falls within the range of r = .24 to r = .41 (p < .001). This weak-to-moderate positive correlation provides structural evidence that teachers who accept responsibility for positive learning outcomes do not automatically accept responsibility for learning failures, empirically confirming the necessity of treating them as separate constructs.

Convergent and Concurrent Validity

Convergent validity is supported by statistically significant associations between RSA scores and established teacher efficacy measures:

• Teacher Efficacy Scales: Guskey (1988) demonstrated that RSA Total R scores correlate positively with the RAND Corporation teacher efficacy measures (r = .38 to .46, p < .01) and the Ashton and Webb Efficacy Scales. Teachers with higher R+ and R- scores demonstrate significantly higher Personal Teaching Efficacy (PTE).
• Mastery Learning Implementation: In longitudinal evaluations, Guskey (1982, 1984) showed that teachers participating in professional development programs focused on Mastery Learning exhibited statistically significant increases in R+ and R- scores relative to control cohorts, with changes in RSA scores corresponding to observable improvements in student achievement.
• Teacher Stress and Job Satisfaction: Subsequent investigations by Woolfolk and Hoy documented meaningful negative associations between low RSA scores and teacher alienation, external locus of work control, and heightened instructional burnout.

Discriminant Validity

Discriminant validity assessments confirm that the RSA does not merely replicate general personality traits or global intelligence. Correlations between RSA scores and measures of social desirability bias (e.g., the Crowne-Marlowe Social Desirability Scale) are routinely nonsignificant (typically r < .15), demonstrating that forced-choice percentage allocation mitigates acquiescence and positive presentation biases.

8. Reliability

The psychometric reliability of the Responsibility for Student Achievement questionnaire has been established across multiple independent cohorts of in-service elementary, middle, and high school teachers, as well as pre-service teacher candidates.

Internal Consistency Reliability

Because the RSA uses an ipsative constant-sum format across paired statements, traditional internal consistency estimates require careful calculation. Studies using item-to-subscale correlations and classical split-half or alpha techniques demonstrate robust reliability parameters:

• Responsibility for Success (R+): Cronbach’s alpha coefficients consistently range between α = .78 and α = .85 across empirical studies (Guskey, 1981, 1988; Woolfolk & Hoy, 1990).
• Responsibility for Failure (R-): Cronbach’s alpha coefficients range from α = .72 to α = .81. The slightly lower alpha observed for R- mirrors broader attributional literature, reflecting greater variance and defensive cognitive processing when assessing negative instructional events.
• Total Composite Responsibility (Total R): Internal reliability for the aggregated 30-item composite routinely exceeds α = .82, reaching up to α = .87 in large-sample demographic surveys.

Test-Retest Stability

Temporal stability assessments over short- and medium-term intervals demonstrate that the RSA captures stable cognitive traits while remaining sensitive to systemic educational training interventions:

• A two-week test-retest assessment across a cohort of urban public school teachers yielded stability coefficients of r = .83 for R+, r = .79 for R-, and r = .84 for Total R (Guskey, 1981).
• Over extended multi-month intervals without targeted intervention, test-retest coefficients remain stable (r ≈ .70), whereas structured mastery learning training causes intentional upward shifts in R- and R+ values, confirming both trait stability and experimental responsiveness.

9. Factor Analysis

The factor structure of the RSA has been subjected to extensive exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) to verify its latent organization.

Exploratory Factor Solutions

Initial principal components analyses using varimax orthogonal and promax oblique rotations by Guskey (1981) confirmed two primary orthogonal factors accounting for the majority of the common variance:

• Factor 1: Academic Success Causality (R+): Defined by robust loadings (> .45) from items delineating student conceptual mastery, high examination performance, rapid learning, and parental commendations tied to teacher pedagogical behavior.
• Factor 2: Academic Failure Causality (R-): Characterized by primary loadings (> .40) from items involving student conceptual difficulty, classroom disengagement, poor test execution, and critical parent/administrative reviews attributed to teacher instructional adjustments.

Confirmatory Factor Modeling and Fit Indices

Subsequent psychometric examinations using modern structural equation modeling (SEM) confirm that a two-factor correlated model provides a markedly superior fit compared to a single-factor unidimensional locus of control model. Fit indices across standard validation samples demonstrate strong structural convergence:

• Root Mean Square Error of Approximation (RMSEA): Values consistently fall between .045 and .058, indicating excellent model fit.
• Comparative Fit Index (CFI): Ranges between .91 and .95 across diverse educator cohorts.
• Standardized Root Mean Square Residual (SRMR): Observed values remain below .06.

Factor loading matrices illustrate that items such as Item 1 (“natural ability vs. encouragement offered”), Item 6 (“helped overcome learning difficulties vs. students usually do well”), and Item 16 (“did not work hard vs. did not provide proper motivation”) manifest standard factor saturation coefficients between .51 and .74 on their targeted latent constructs without substantial cross-loadings.

10. Instrument / Measurement Tool

• Instrument Name: Responsibility for Student Achievement Questionnaire (RSA)
• Developer: Dr. Thomas R. Guskey (1981, 1982, 1988)
• Construct Assessed: Teacher attributional locus of causality and perceived personal responsibility for student academic success and failure
• Number of Items: 30 dyadic items (each presenting two alternative causal explanations)
• Item Categorization:
  • R+ Items (Success Events): Items 1, 3, 5, 6, 7, 10, 13, 15, 18, 19, 21, 22, 23, 26, 27 (15 items)
  • R- Items (Failure Events): Items 2, 4, 8, 9, 11, 12, 14, 16, 17, 20, 24, 25, 28, 29, 30 (15 items)
• Response Format: Constant-sum percentage allocation (ipsative forced-choice). For each item, respondents allocate a weight or percentage to Option (a) and Option (b) reflecting their degree of endorsement. The sum of (a) and (b) for each item must always equal exactly 100% (e.g., 85% – 15%, 50% – 50%, or 99% – 1%).
• Scoring Architecture:
  • Each item pairs an internal attribution option (identifying the teacher’s instruction, planning, effort, or motivation) with an external attribution option (identifying student ability, innate traits, parent factors, or testing conditions).
  • The internal choice is marked as either R+ (internal attribution for success) or R- (internal attribution for failure).
  • R+ Subscale Score: Calculate the mean percentage allocated to the internal teacher-responsible alternative across the 15 success items.
  • R- Subscale Score: Calculate the mean percentage allocated to the internal teacher-responsible alternative across the 15 failure items.
  • Total R Score: The overall mean percentage allocated to internal options across all 30 items: Total R = (R+ + R-) / 2.
• Target Population: In-service teachers across all K–12 instructional domains, pre-service teacher candidates, school administrators, and higher education instructional staff.
• Estimated Completion Time: 15 to 25 minutes.

11. Permissions & Fee and Test Year

The Responsibility for Student Achievement (RSA) questionnaire was formally published in the peer-reviewed psychometric literature in 1981, with expanded empirical validation studies appearing in 1982 and 1988. The scale was established as an open-access empirical instrument for scholarly, educational, non-commercial, and clinical research purposes.

Educators, psychometricians, and university researchers are permitted to utilize, administer, and reproduce the instrument for non-commercial research investigations provided proper scholarly citation and attribution are given to Dr. Thomas R. Guskey and the original journal publications. The instrument and associated scoring protocols have been widely preserved and distributed through academic resource repositories, notably including the institutional measurement archives maintained by Dr. Anita Woolfolk Hoy at The Ohio State University. Commercial adaptations, incorporation into fee-for-service enterprise software platforms, or third-party distribution for profit require formal permission from the copyright holders and the primary author.

12. References

• Bandura, A. (1997). Self-efficacy: The exercise of control. W. H. Freeman and Company.
• Crandall, V. C., Katkovsky, W., & Crandall, V. J. (1965). Children’s beliefs in their own control of reinforcements in intellectual-academic achievement situations. Child Development, 36(1), 91–109. https://doi.org/10.2307/1126628
• Guskey, T. R. (1981). Measurement of the responsibility teachers assume for academic successes and failures in the classroom. Journal of Teacher Education, 32(3), 44–51. https://doi.org/10.1177/002248718103200310
• Guskey, T. R. (1982). The effects of change in instructional effectiveness on teacher attitudes and beliefs. The Journal of Educational Research, 76(2), 110–114. https://doi.org/10.1080/00220671.1982.10885434
• Guskey, T. R. (1988). Teacher efficacy, self-concept, and attitudes toward the implementation of instructional innovation. Teaching and Teacher Education, 4(1), 63–69. https://doi.org/10.1016/0742-051X(88)90025-X
• Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80(1), 1–28. https://doi.org/10.1037/h0092976
• Weiner, B. (1979). A theory of motivation for some classroom experiences. Journal of Educational Psychology, 71(1), 3–25. https://doi.org/10.1037/0022-0663.71.1.3
• Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92(4), 548–573. https://doi.org/10.1037/0033-295X.92.4.548
• Woolfolk, A. E., & Hoy, W. K. (1990). Prospective teachers’ sense of efficacy and beliefs about control. Journal of Educational Psychology, 82(1), 81–91. https://doi.org/10.1037/0022-0663.82.1.81

13. Items of the Scale (Questionnaire)