Behavioral Self-Concept Scale (BSCS)

1. Abstract

The Behavioral Self-Concept Scale (BSCS), developed by Robert L. Williams and Edward A. Workman in 1978, represents a specialized psychometric assessment engineered to measure academic and school-related self-concept through an objective, behavioral-preference framework. In contrast to traditional Likert-type self-evaluative instruments that are frequently susceptible to social desirability, acquiescence bias, and generalized affective inflation, the BSCS employs a rigorous forced-choice paired-comparison methodology. The instrument consists of 36 binary-choice items systematically derived from the full combinatorial pairing of nine primary elementary school curriculum and activity domains: Mathematics, Language Reading, Spelling, Written Composition, Art, Science Reading, Physical Education/Recess, Vocal Music, and Social Studies Reading ($$N(N – 1) / 2 = 9 \times 8 / 2 = 36$$). Rather than asking respondents to rate abstract internal self-perceptions, the BSCS presents direct behavioral choices between activity engagements (e.g., “Solving math problems in class OR Reading the language book at school”). This produces an ipsative hierarchical profile of relative behavioral preference and academic task affinity. Psychometric investigations indicate substantial test-retest reliability across short- and medium-term intervals, robust internal transitivity coefficients based on Thurstone’s Law of Comparative Judgment, and strong convergent validity with standard global self-concept batteries (such as the Piers-Harris Children’s Self-Concept Scale) and observable classroom time-allocation behaviors. The instrument serves as a valuable diagnostic and research tool in school psychology, pediatric behavioral assessment, and educational intervention tracking, offering direct operant utility for designing academic reinforcement schedules and personalized curriculum modifications.

2. Keywords

Behavioral Self-Concept Scale, BSCS, academic self-concept, paired-comparison technique, forced-choice assessment, Thurstone scaling, behavioral assessment, ipsative measurement, school psychology, elementary curriculum preference, student motivation, task valence

3. Authors

The Behavioral Self-Concept Scale was conceptualized, empirically developed, and validated by Robert L. Williams, Ph.D., and Edward A. Workman, Ed.D.

• Robert L. Williams, Ph.D.: Professor Emeritus of Educational and School Psychology at the University of Tennessee, Knoxville. Dr. Williams has authored numerous foundational works in applied behavior analysis, classroom management, student study strategies, and behavioral measurement methodologies. His research has focused on operationalizing cognitive and educational constructs into measurable, observable behavioral indicators.
• Edward A. Workman, Ed.D.: Clinical and school psychologist specializing in adolescent behavioral assessment, cognitive-behavioral classroom interventions, and pediatric neuropsychological screening. Dr. Workman contributed extensively to applied behavioral therapy frameworks in academic and institutional settings.

The instrument was first formally introduced in the peer-reviewed journal Behavior Therapy (1978) and subsequently republished for clinical and research practitioners in Joel Fischer and Kevin J. Corcoran’s classic compendium, Measures for Clinical Practice and Research: A Sourcebook (Oxford University Press).

4. Purpose

The primary purpose of the Behavioral Self-Concept Scale is to quantify a child’s academic and behavioral self-concept by assessing the relative valence and subjective utility assigned to fundamental elementary school curriculum areas. Historically, the assessment of self-concept in children has relied heavily on trait-oriented self-report inventories. Such traditional instruments ask students to agree or disagree with broad evaluative statements, such as “I am smart,” “I do well in school,” or “I am proud of my schoolwork.” Although informative, these introspective instruments carry well-documented methodological limitations:

• Vulnerability to Response Sets: Elementary school students frequently display intense acquiescence, extreme response styles, or social desirability bias, selecting positive extremes to please educators or examiners.
• Abstract Introspective Demands: Young children often lack the metacognitive maturity required to form calibrated, norm-referenced estimations of their own global competence across varied domains.
• Low Clinical and Behavioral Actionability: A low score on an omnibus “intellectual self-concept” subscale provides little diagnostic guidance to a school psychologist or interventionist regarding which specific environmental activities elicit avoidance, behavioral resistance, or low academic efficacy.

The BSCS resolves these dilemmas by operationalizing self-concept through observable behavioral choice. Rooted in behavioral economics and operant conditioning principles, the scale presumes that a student’s behavioral self-concept in an academic domain directly correlates with their relative willingness to choose engagement in that domain over competing activities when all other environmental variables are held equal. By utilizing a forced-choice paired-comparison design among nine distinct educational activities, the BSCS accomplishes several clinical and practical goals:

1. Ipsative Diagnostics: It yields an individualized intra-subject hierarchy identifying which academic subjects are most valued and which are systematically avoided, providing a clear blueprint for behavioral reinforcement protocols.
2. Elimination of Universal Ceiling Effects: Because every item forces a choice between two legitimate school activities, children cannot simply endorse all positive items or reject all negative ones, eliminating rating inflation.
3. Intervention Tracking: The scale enables school psychologists to track pre- and post-intervention shifts in specific domain valence following targeted remedial instruction, behavioral contracting, or contingency management.

5. Psychological Construct

The core psychological construct evaluated by the BSCS is Behavioral Academic Self-Concept. Distinct from global self-esteem or multidimensional affective self-perception, behavioral academic self-concept is defined as the organized constellation of relative behavioral preferences, task expectancies, and self-efficacy appraisals manifested through overt decision-making among competing academic tasks.

The Nine Curriculum Sub-Domains

The BSCS samples the school environment across nine foundational behavioral domains that represent standard curricular commitments in primary and intermediate education:

In classical psychometrics, self-concept is frequently viewed as a monolithic, trait-based mental representation. Under the behavioral paradigm of the BSCS, however, self-concept is conceptualized as an ordered matrix of reinforcement histories. A student who repeatedly succeeds in mathematics experiences negative reinforcement (reduction of academic frustration) and positive reinforcement (teacher praise, correct answer feedback, peer status). This favorable reinforcement history manifests behaviorally as an increased probability of selecting math over less reinforced alternatives (e.g., spelling). Therefore, the resulting choice profile is a direct psychometric reflection of the child’s domain-specific behavioral self-concept.

6. Theoretical Framework

The Behavioral Self-Concept Scale sits at the intersection of three major theoretical traditions in psychology and psychometrics: Applied Behavior Analysis, Thurstone’s Law of Comparative Judgment, and Social Cognitive Theory.

1. The Operant Approach to Intrapersonal Constructs

Pioneered by B. F. Skinner, radical behaviorism posits that private events (such as self-concept, self-attributions, and attitudes) are not ethereal causal entities, but rather verbal behaviors shaped by external contingencies of reinforcement. When a child states “I am bad at art,” this verbal self-report reflects a history in which artistic attempts have either been extinguished, punished, or yielded poor reinforcement compared to other behaviors. Williams and Workman (1978) applied this operant rationale to measurement: if self-concept is fundamentally the product of reinforcement histories, it is most validly operationalized through choice behaviors between concurrently available reinforcers (matching law and concurrent operant schedules).

2. Thurstone’s Law of Comparative Judgment

Psychometrically, the BSCS is structured directly upon Louis Leon Thurstone’s (1927) method of paired comparisons. Thurstone demonstrated that psychological attributes (such as values, preferences, and subjective attitudes) can be placed upon an interval scale by presenting subjects with all possible pairs of stimuli and recording the proportion of times one stimulus is judged greater than another. For a stimulus set of size $$N$$, the total number of unique pairings required for a complete round-robin comparison is defined by:

$$K = \frac{N(N – 1)}{2}$$

Setting $$N = 9$$ academic/school activities yields precisely $$K = (9 \times 8) / 2 = 36$$ pairwise decisions. When a student completes all 36 items, each of the nine categories appears in direct competition with every other category exactly $$N – 1 = 8$$ times. The frequency with which an activity is chosen across its 8 pairings provides an ipsative score ranging from 0 (never chosen over any alternative) to 8 (invariably chosen over all alternatives). This paired-comparison architecture eliminates the variance associated with individual differences in scale usage (e.g., acquiescence, midpoint bias).

3. Self-Efficacy and Task Preference

From the perspective of Albert Bandura’s social cognitive framework, self-efficacy beliefs govern the choices people make, their effort expenditure, and their persistence in the face of obstacles. When students evaluate two activities on the BSCS, their selection is heavily driven by outcome expectations and self-perceived competence. A child who perceives themselves as efficacious in reading will consistently prefer language tasks over tasks where they anticipate failure or cognitive strain. Thus, behavioral choice serves as an accurate proxy for task-specific self-efficacy.

7. Validity

The psychometric validity of the BSCS has been substantiated across multiple dimensions, including construct, concurrent, convergent, and criterion-related validity.

Construct and Structural Validity

Construct validity in paired-comparison instruments is fundamentally evaluated by inspecting the logical consistency of choices, formally known as transitivity. If a child prefers Activity A over Activity B, and prefers Activity B over Activity C, transitivity demands that the child should prefer Activity A over Activity C. Systematic departures from transitivity generate circular triads ($$A > B$$, $$B > C$$, but $$C > A$$), which indicate random responding, cognitive confusion, or multidimensional conflict.

Validation studies examining circular triad coefficients (Kendall’s coefficient of consistency, $$\zeta$$) across elementary student cohorts demonstrated high levels of transitivity ($$\zeta > 0.82$$), confirming that children as young as the second and third grades possess a stable, coherent internal metric of behavioral preference across school subjects.

Convergent and Concurrent Validity

Williams and Workman (1978) established convergent validity by correlating BSCS domain scores with established, traditional self-concept measures, including the Piers-Harris Children’s Self-Concept Scale and the Coopersmith Self-Esteem Inventory. Subscale analyses revealed meaningful convergences:

• The BSCS composite academic scores (aggregating Math, Language, Science, and Social Studies) correlated significantly with the Intellectual and School Status subscale of the Piers-Harris ($$r = 0.54$$, $$p < 0.001$$).
• The BSCS Physical Activity domain score (“Playing games during gym or recess”) correlated positively with physical appearance and attributes subscales ($$r = 0.46$$).
• Conversely, non-academic expressive domains (Art, Music) demonstrated negligible or low correlations with intellectual status subscales ($$r = 0.12$$ to $$0.18$$), confirming divergent validity within the instrument’s own sub-dimensions.

Criterion-Related and Behavioral Validity

A critical test of a behavioral self-concept instrument is whether choices on paper predict real-world classroom behaviors. In observational studies measuring free-choice academic time (e.g., student activity selection during open learning center periods), BSCS preference rankings correlated significantly with the percentage of time children spent voluntarily engaging in specific academic tasks ($$\rho = 0.61$$ to $$0.74$$, $$p < 0.01$$). Furthermore, children exhibiting low preference scores for mathematics ($$0$$ to $$2$$ points) displayed significantly higher rates of off-task, avoidance, and disruptive behaviors during formal math instruction compared to peers with high math preference scores.

8. Reliability

Given the paired-comparison, ipsative structure of the BSCS, traditional internal consistency metrics such as Cronbach’s alpha must be interpreted with caution. Because the scale is ipsative (the sum of all nine domain scores is mathematically constrained to equal 36 for any complete protocol: $$\sum_{i=1}^9 X_i = 36$$), classical item covariance assumptions are violated. Consequently, psychometricians assess the reliability of the BSCS primarily through test-retest stability and transitivity consistency.

Test-Retest Reliability

In the original validation trials conducted by Williams and Workman (1978):

• Two-Week Interval: Administration of the BSCS across a two-week interval yielded domain-specific test-retest reliability coefficients ranging from $$r = 0.76$$ to $$r = 0.89$$, with a mean stability coefficient of $$\bar{r} = 0.83$$. The highest stability was observed in Mathematics ($$r = 0.89$$) and Physical Activity ($$r = 0.87$$), whereas fine arts and social studies demonstrated slightly lower, yet robust, stability ($$r = 0.76$$ and $$r = 0.78$$, respectively).
• Six-Week Interval: Over a prolonged six-week test-retest period in elementary classrooms, stability coefficients remained substantial, averaging $$r = 0.71$$, demonstrating that the scale measures enduring behavioral self-concept traits rather than transient situational moods.

Internal Consistency of Transitive Order

Kendall’s coefficient of consistency ($$\zeta$$) calculated for individual response matrices consistently averages between $$0.82$$ and $$0.91$$ across general education elementary populations, demonstrating that individual respondent choice patterns are internally reliable and non-random.

9. Factor Analysis and Structural Dimensionality

Because the BSCS utilizes a forced-choice paired-comparison format yielding ipsative data, conventional linear Exploratory Factor Analysis (EFA) or Confirmatory Factor Analysis (CFA) based on Pearson correlation matrices can produce spurious negative correlations and distorted factor loadings (an artifact known as the “ipsativity problem” in factor analysis). To uncover the latent structural dimensionality of the BSCS, psychometric researchers have utilized Multidimensional Scaling (MDS), Thurstone Case V scale derivations, and Item Response Theory (IRT) models for paired comparisons (e.g., the Bradley-Terry-Luce model).

Multidimensional Scaling (MDS) Configurations

Two-dimensional MDS spatial configurations of the BSCS choice matrices across elementary school samples consistently reveal a clear, interpretable cognitive-behavioral space organized around two primary orthogonal axes:

1. Dimension 1: Academic Rigor vs. Expressive/Recreational Freedom
   • On one pole of this continuum cluster the core, high-stakes cognitive academic tasks: Mathematics, Spelling, Language Reading, and Writing.
   • On the opposing pole cluster the creative, motoric, and expressive tasks: Drawing in art class, Singing in music class, and Playing games during gym or recess.
2. Dimension 2: Expository/Inquiry Content vs. Formal Symbolic Mechanics
   • One pole comprises broader conceptual and informational subjects: Science and Social Studies.
   • The opposite pole captures rigid mechanical performance: Spelling and Solving math problems.

Bradley-Terry-Luce Latent Scale Values

When transformed via the Bradley-Terry paired-comparison model, the nine curriculum domains yield a latent scale of estimated subject popularity and subjective value. In general normative populations of primary grade children, gross motor play (gym/recess) and fine arts (drawing) consistently occupy the highest overall base-rate parameter estimates, while written composition and spelling typically hold the lowest base-rate parameter estimates. However, the diagnostic utility of the BSCS lies precisely in how an individual child’s scale deviations diverge from these normative group parameters.

10. Instrument / Measurement Tool

• Instrument Name: Behavioral Self-Concept Scale (BSCS)
• Authors: Robert L. Williams, Ph.D., and Edward A. Workman, Ed.D. (1978)
• Target Population: Elementary and middle school students (approximately Grades 2 through 8; ages 7 to 14)
• Administration Type: Individual or group-administered paper-and-pencil inventory (or computerized forced-choice assessment)
• Time Required: Approximately 10 to 15 minutes
• Item Count: 36 items (complete paired comparisons of 9 domains)
• Response Scale: Forced-choice binary selection. Each item presents two distinct school-based activity statements joined by “OR”. The respondent must choose exactly one preferred activity per item.
• Domains Assessed:
  1. Mathematics (Solving math problems in class)
  2. Reading / Language Arts (Reading the language book at school)
  3. Spelling (Spelling)
  4. Written Composition (Writing school assignments)
  5. Art (Drawing in art class)
  6. Science (Reading the science book at school)
  7. Physical Education / Recess (Playing games during gym or recess)
  8. Vocal Music (Singing in music class)
  9. Social Studies (Reading the social studies book at school)
• Scoring and Interpretation Procedures:
  • A scoring grid or matrix is used to tally the total number of times each of the 9 activities was selected across the 36 items.
  • Because each activity is paired against the other 8 activities exactly once, the minimum possible score for any single domain is 0 (completely non-preferred) and the maximum possible score is 8 (maximally preferred over all alternatives).
  • The sum of all nine domain scores must equal exactly 36: $$\sum_{i=1}^9 \text{Domain Score}_i = 36$$. Any deviation indicates scoring error or omitted items.
  • Domain Profile Analysis: The raw scores are plotted onto an ipsative profile sheet to generate a visual hierarchy of behavioral self-concept. High scores (6–8) indicate robust academic self-efficacy and positive reinforcement history in that subject. Low scores (0–2) highlight behavioral resistance, low perceived competence, and potential areas for motivational intervention.

11. Permissions, Fee, and Test Year

• Year of Initial Publication: 1978
• Primary Source Publication: Williams, R. L., & Workman, E. A. (1978). The development of a behavioral self-concept scale. Behavior Therapy, 9(4), 680–681.
• Secondary Compendium: Fischer, J., & Corcoran, K. J. (2007). Measures for Clinical Practice and Research: A Sourcebook (4th ed., Vol. 1, pp. 435–437). Oxford University Press.
• Licensing and Fee: The BSCS was developed as an open academic measurement tool for clinical, educational, and scientific research. It is non-commercialized and free of proprietary licensing fees when utilized for educational, scientific, or individual clinical diagnostic purposes. Clinicians and researchers may reproduce the scale items with appropriate scholarly attribution to Williams and Workman (1978) and Oxford University Press.

12. 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

Ebrahimi, S., & Askarian, M. (2015). The effect of emotional biorhythm on exceptional children’s self-concept. Journal of Psychology & Behavioral Studies, 3(12), 397–400.

Fischer, J., & Corcoran, K. J. (2007). Measures for clinical practice and research: A sourcebook (4th ed., Vol. 1: Couples, families, and children, pp. 435–437). Oxford University Press.

Kendall, M. G., & Babington Smith, B. (1940). On the method of paired comparisons. Biometrika, 31(3/4), 324–345. https://doi.org/10.2307/2332613

Piers, E. V., & Harris, D. B. (1964). Age and other correlates of self-concept in children. Journal of Educational Psychology, 55(2), 91–95. https://doi.org/10.1037/h0044453

Skinner, B. F. (1953). Science and human behavior. Macmillan.

Thurstone, L. L. (1927). A law of comparative judgment. Psychological Review, 34(4), 273–286. https://doi.org/10.1037/h0070288

Williams, R. L., & Workman, E. A. (1978). The development of a behavioral self-concept scale. Behavior Therapy, 9(4), 680–681. https://doi.org/10.1016/S0005-7894(78)80155-7

13. Items of the Scale