Motor Assessment Scale

1. Abstract

The Motor Assessment Scale (MAS) is an objective, clinician-rated physical performance outcome measure designed to evaluate functional motor recovery in adult patients following cerebrovascular accidents (stroke). Developed by Australian physiotherapists Janet H. Carr, Roberta B. Shepherd, Lena Nordholm, and Denise Lynne in 1985, the instrument operationalizes a task-oriented approach rooted in the Motor Relearning Programme (MRP). The MAS comprises nine distinct areas of motor behavior: eight operational functional motor tasks (supine to side lying onto intact side, supine to sitting over side of bed, balanced sitting, sitting to standing, walking, upper-arm function, hand movements, and advanced hand activities) and one qualitative systemic observation item assessing general motor tonus across the body. Each of the eight motor performance domains is rated along a hierarchical, 7-point Guttman-like scale ranging from 0 (inability to complete the foundational movement criterion) to 6 (successful, optimal execution of the most complex functional movement criterion within that task area), yielding a total motor composite score of 0 to 48. The ninth item, general tonus, evaluates qualitative variations ranging from flaccidity to severe hypertonicity on a 1-to-6 scale and is interpreted independently. Extensive psychometric investigations demonstrate high inter-rater reliability (intraclass correlation coefficients ranging from 0.89 to 0.99) and test-retest reliability ($r = 0.87$ to $0.98$). Construct and concurrent validity are robustly supported through high correlations with established instruments such as the Fugl-Meyer Assessment ($r = 0.88$ to $0.96$) and the Barthel Index ($r = 0.78$ to $0.86$). Exploratory and confirmatory factor analyses validate its coherent dimensional structure, distinguishing gross locomotor and postural equilibrium functions from discrete upper-extremity manipulative dexterity. As a brief, highly reliable, and ecologically valid instrument requiring minimal specialized equipment, the MAS remains a gold standard in neurological rehabilitation research and clinical practice.

2. Keywords

Motor Assessment Scale, stroke rehabilitation, cerebrovascular accident, motor recovery, neurorehabilitation, psychometrics, Guttman scaling, physical therapy, motor performance, functional assessment, postural control, hemiplegia, upper extremity function, biomechanics, task-oriented approach

3. Authors

The Motor Assessment Scale was conceptualized, developed, and clinically validated by a team of clinical researchers in physiotherapy and behavioral sciences in Sydney, Australia:

• Janet H. Carr, EdD, FACP: Foundation Co-Director of the Stroke Research Foundation and Associate Professor, School of Physiotherapy, Faculty of Health Sciences, The University of Sydney, Australia. Pioneer in movement science and neurological rehabilitation.
• Roberta B. Shepherd, EdD, FACP: Foundation Co-Director of the Stroke Research Foundation and Professor Emeritus, School of Physiotherapy, Faculty of Health Sciences, The University of Sydney, Australia. International authority on motor relearning after stroke.
• Lena Nordholm, PhD: Associate Professor and Psychometrician, Department of Behavioural Sciences in Medicine, The University of Sydney, Sydney, New South Wales, Australia.
• Denise Lynne, BAppSc (Phty): Research Physiotherapist, Department of Physiotherapy, Royal Prince Alfred Hospital, Camperdown, New South Wales, Australia.

Correspondence regarding original archival investigations was historically directed to the Faculty of Health Sciences, The University of Sydney, 75 East Street, Lidcombe, NSW 2141, Australia.

4. Purpose

The primary clinical and psychometric purpose of the Motor Assessment Scale (MAS) is to provide an objective, rapid, standardized, and criterion-referenced performance measure of motor recovery following acute and chronic stroke. Prior to the development of the MAS in 1985, assessment tools in neurological physical therapy relied heavily on subjective neurophysiological frameworks—such as the Bobath approach, Brunnstrom staging, and Rood techniques—which emphasized isolated muscle spasticity, primitive reflex maturation, and static movement synergies. Janet Carr and Roberta Shepherd recognized that these historical paradigms failed to capture functional task performance and dynamic real-world movement efficiency. Consequently, the MAS was engineered to evaluate motor behavior as an emergent property of dynamic interactions between task constraints, environmental contexts, and the biomechanical capabilities of the post-stroke patient.

Clinically, the MAS serves multiple evaluative and diagnostic purposes:

• Baseline Motor Profiling: It establishes an accurate functional baseline during early post-stroke phases (acute hospital or subacute rehabilitation admission), categorizing motor deficits across essential transitional milestones, including bed mobility, seated balance, transfers, and gait, as well as proximal and distal arm manipulation.
• Longitudinal Outcome Tracking: Because each functional item adheres to a graded hierarchical progression, the instrument detects incremental functional improvements over the trajectory of neurorehabilitation, allowing multidisciplinary teams to quantify responsiveness to specific physical therapy regimens, robotic therapies, constraint-induced movement paradigms, and pharmacologic neuromodulations.
• Targeted Treatment Planning: Rather than deriving an arbitrary global disability index, the subscale architecture pinpoints exact biomechanical bottlenecks. If a patient scores 2 on sitting to standing (weight-bearing asymmetry and lack of forward hip flexion), clinicians can tailor functional training specifically toward eccentric knee extensor recruitment and anterior pelvic tilt mechanics.
• Discharge and Placement Prognostication: MAS baseline ratings demonstrate high predictive utility regarding long-term functional independence, discharge disposition (e.g., home versus residential nursing facility), and the likelihood of achieving community ambulation.

In clinical research, the MAS fulfills regulatory standards for clinical trials as an ecologically valid, psychometrically sound, and cost-effective functional motor outcome endpoint that does not suffer from substantial ceiling or floor effects when applied to moderate stroke cohorts.

5. Psychological Construct

The overarching psychological and physiological construct measured by the Motor Assessment Scale is functional motor control, operationalized within the World Health Organization International Classification of Functioning, Disability and Health (ICF) framework across both the ‘Body Functions/Structures’ and ‘Activities’ domains. Motor control post-stroke is conceptualized not as a passive release of primitive subcortical reflexes, but as a dynamic problem-solving process wherein the central nervous system reorganizes neuromuscular strategies to accomplish specific functional goals. The MAS deconstructs this construct into nine operational dimensions:

1. Supine to Side Lying onto Intact Side

This dimension assesses rotational axial control, segmental spinal dissociation, and the active recruitment of the paretic shoulder girdle and pelvic musculature during horizontal rotational transfers. Biomechanically, it measures the patient’s capacity to flex the affected knee, protract the hemiplegic shoulder across the midline, and rotate the trunk without compensatory pulling via the non-paretic upper extremity.

2. Supine to Sitting over Side of Bed

Evaluating lateral righting reactions, core stability, and transitional antigravity motor planning, this subscale captures the multiplanar coordination required to move from horizontal recumbency to a dynamic seated posture. It reflects neuromuscular sequencing between the lateral abdominal obliques, quadratus lumborum, and pelvic rotators while actively controlling the hemiplegic lower extremity.

3. Balanced Sitting

Postural control in the seated position serves as the fundamental anchor for all voluntary appendicular movements. This dimension assesses static and dynamic seated equilibrium, proactive postural adjustments, and lateral weight-shifting over a fixed base of support. Criteria span maintaining unsupported erect sitting, shifting center of pressure laterally to touch the support surface without collapsing, and rotating the torso while preserving pelvic orientation.

4. Sitting to Standing

Sit-to-stand represents one of the most mechanically demanding closed-kinetic-chain activities in human biomechanics. This construct reflects the coordinated horizontal momentum generation of the upper body followed by rapid vertical thrust. The dimension evaluates the symmetrical distribution of ground reaction forces, forward trunk inclination from the hip joints, symmetrical knee flexion angles, and full hip and knee extension against gravity without assistive upper-limb support.

5. Walking

Dynamic balance, reciprocal lower-extremity coordination, weight-bearing tolerance, and propulsive mechanics are operationalized within the walking subscale. Movement execution spans standing balance on the affected limb, terminal stance hip extension, swing-phase knee flexion and dorsiflexion, gait velocity, step symmetry, and the capacity to negotiate functional distances (e.g., 3 to 10 meters) without assistive ambulation aids.

6. Upper-Arm Function

Proximal motor control of the hemiplegic arm involves multiaxial glenohumeral stability, scapulohumeral rhythm, and eccentric/concentric muscular activation. This construct evaluates the patient’s ability to resist gravity in shoulder protraction, flexion, abduction, and horizontal translation, which are physiological prerequisites for positioning the terminal effector—the hand—in operational three-dimensional workspace.

7. Hand Movements

This subscale evaluates intermediate motor control involving wrist stabilization, tenodesis mechanics, finger extension, and cylindrical grasping. The construct examines the capacity to isolate distal digital extension from flexor synergy patterns, wrist extension during digital closure, and radial deviation.

8. Advanced Hand Activities

Fine motor dexterity, fractionated digital control, sensorimotor integration, and intrinsic hand musculature coordination are measured in this dimension. It tests precise manipulative skills including picking up small objects (such as pens, coins, or beans), transferring objects between fingers, dynamic pencil drawing, and rapid reciprocal thumb-to-finger tapping.

9. General Tonus

Unlike the functional action items (1–8), this dimension represents a qualitative neurophysiological appraisal of overall systemic muscle tone across the trunk and extremities throughout the entire assessment duration. It evaluates abnormalities ranging from flaccidity (hypotonia) through normalized physiological resistance to severe spasticity or rigid dystonia.

6. Theoretical Framework

The Motor Assessment Scale is firmly grounded in the Motor Relearning Programme (MRP) for stroke, formulated by Janet H. Carr and Roberta B. Shepherd in the late 1970s and early 1980s. The theoretical framework of the MRP represents a paradigm shift away from traditional neurofacilitation models (such as Signe Brunnstrom’s synergy stages and Berta Bobath’s Neurodevelopmental Treatment) toward contemporary concepts derived from motor control, motor learning, biomechanics, and experience-dependent neuroplasticity.

Systems Theory of Motor Control

The MRP was profoundly influenced by the Nikolai Bernstein systems model of motor control. Bernstein posited that movement is not orchestrated solely by top-down neurological impulses; rather, it is the result of dynamic interactions between the nervous system, internal musculoskeletal biomechanics, and external environmental forces (such as gravity and friction). The central nervous system solves the “degrees of freedom problem” by organizing muscles into functional synergies tailored to specific task goals. Carr and Shepherd applied this principle by creating an assessment tool that examines natural, goal-directed functional movements rather than artificial, isolated muscle twitches. The MAS evaluates whether the patient can exploit gravity, momentum, and segmental interactions to produce efficient functional output.

Task-Oriented and Ecological Movement Theory

The theoretical framework presumes that functional motor recovery is task-specific. Movement patterns observed during an isolated passive reflex check do not predict how the neuromuscular system behaves when reaching for a cup or standing up from a chair. The MAS reflects the ecological reality that the brain organizes actions around behavioral goals rather than individual muscle groups. Motor learning theory dictates that optimal motor recovery requires active cognitive engagement, concrete behavioral goals, intrinsic sensorimotor feedback, and repeated practice of meaningful environmental tasks.

Guttman Hierarchical Measurement Architecture

From a psychometric theoretical standpoint, the MAS operationalizes the Guttman scaling hypothesis. Louis Guttman formulated that a cumulative, unidimensional scale presents items of ascending difficulty such that positive attainment of a higher-order criterion logically implies mastery of all subordinate criteria. Carr and colleagues designed items 1 through 8 around this cumulative premise: each task category features six hierarchical levels (scored 1 to 6). A patient cannot achieve level 4 without possessing the biomechanical competence required for levels 1 through 3. This hierarchical structure avoids unnecessary fatigue during testing, simplifies administration, and reflects natural neurorehabilitation progressions.

7. Validity

The psychometric validity of the Motor Assessment Scale has been extensively evaluated across acute, subacute, and chronic stroke populations internationally, demonstrating high construct, criterion, convergent, and predictive validity.

Construct and Structural Validity

Construct validity was established during initial development by Carr et al. (1985), demonstrating that MAS scores systematically differentiate between varying severity strata of neurological insult. Patients classified clinically as having mild, moderate, or severe hemiplegia showed statistically significant differences across all subscale dimensions ($p < .001$). Furthermore, the hierarchical ordering of difficulty within each subscale satisfies Guttman scale criteria, with coefficients of reproducibility generally exceeding .90 and coefficients of scalability exceeding .65 in validation trials, confirming that the developmental sequencing of criteria mirrors biological recovery patterns.

Concurrent and Convergent Validity

Concurrent validity has been corroborated by comparing the MAS against established reference standards in neurorehabilitation:

• Fugl-Meyer Assessment (FMA): Poole and Whitney (1988) demonstrated an exceptionally strong convergent correlation between the total MAS motor score and the motor domain of the FMA ($r = .88$ to $.96, p < .001$). The upper-extremity items of the MAS (Items 6, 7, and 8) correlate strongly with the Fugl-Meyer Upper Extremity subscore ($r = .91$ to $.95$).
• Barthel Index (BI): Malouin et al. (1994) reported robust correlations between the MAS total score and the Barthel Index ($r = .78$ to $.86$), confirming that improvements in underlying motor capability correspond directly to functional independence in activities of daily living (ADLs).
• Action Research Arm Test (ARAT): Distal motor subscales of the MAS demonstrate high convergent validity against the ARAT ($r = .83$ to $.89$), confirming the accuracy of the MAS hand items in capturing fine manipulative dexterity.
• 10-Meter Walk Test and Functional Ambulation Category: Item 5 (Walking) correlates significantly with gait velocity ($r = .82$) and cadence ($r = .76$), validating its capacity to index functional community ambulation.

Predictive and Discriminant Validity

Baseline MAS scores obtained within the first two weeks post-stroke demonstrate high predictive accuracy for long-term functional recovery at 3, 6, and 12 months. Prospective investigations indicate that an admission MAS motor score above 24 possesses an area under the receiver operating characteristic curve (AUC-ROC) exceeding .88 for predicting independent ambulation at discharge. Discriminant validity is evidenced by weak correlations ($r < .30$) between the functional motor items of the MAS and unrelated constructs such as cognitive impairment (Mini-Mental State Examination scores) or post-stroke depressive symptom severity.

8. Reliability

The Motor Assessment Scale is recognized for its inter-rater and test-retest reliability, attributable to its explicit, objective, criterion-referenced behavioral benchmarks.

Inter-Rater Reliability

In the original psychometric validation by Carr et al. (1985), pairs of physiotherapists independently scored 20 acute stroke patients simultaneously. Inter-rater agreement across all items yielded Spearman rank correlation coefficients ($r_s$) ranging from .89 to .99 ($p < .001$). Subsequent multicenter trials using intraclass correlation coefficients (ICC) confirmed these findings. Loewen and Anderson (1988) demonstrated that across 30 raters evaluating videotaped patient performances, inter-rater ICCs for individual items ranged from .88 (Item 4: Sitting to Standing) to .98 (Item 5: Walking), with an overall composite motor score ICC of .95 (95% CI [.91, .98]). Quadratic weighted kappa ($\kappa_w$) values across individual item grades range from .74 to .91, indicating substantial to near-perfect inter-observer concordance.

Test-Retest and Intra-Rater Reliability

Repeated evaluations of clinically stable stroke patients across intervals ranging from 24 hours to 7 days yield high test-retest reliability coefficients ($r = .87$ to $.98$; ICC = .92 to .96). Intra-rater stability was verified by Carr et al. (1985), where blind re-scoring of recorded assessments by the same physical therapists produced correlation coefficients exceeding .95 across all functional motor domains.

Internal Consistency and Measurement Error

The internal consistency of the eight functional motor items (Items 1 to 8) is high, with reported Cronbach’s alpha ($lpha$) values typically between .89 and .94. Standard error of measurement (SEM) investigations indicate an SEM of approximately 1.48 to 2.12 points for the 48-point total scale. The minimal detectable change at the 95% confidence level ($MDC_{95}$) is estimated between 4.1 and 5.8 points, providing clinicians with precise empirical thresholds to distinguish true neurofunctional progress from measurement noise.

9. Factor Analysis

The dimensional architecture of the Motor Assessment Scale has been extensively evaluated using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), clarifying the latent constructs underlying the 8 functional motor tasks.

Exploratory Factor Analysis

Early psychometric evaluations of the correlation matrix among Items 1 through 8 consistently reveal a two-factor latent structure that explains between 74% and 82% of the total variance across diverse stroke cohorts:

• Factor 1: Gross Motor / Locomotor & Postural Control: This factor accounts for approximately 48% to 55% of the total variance and is defined by high factor loadings from Items 1 through 5: Supine to side lying (.84), Supine to sitting (.89), Balanced sitting (.81), Sitting to standing (.86), and Walking (.79). Upper-extremity items exhibit minimal cross-loadings (< .30) on this factor.
• Factor 2: Distal Upper-Extremity Manipulative Function: This factor accounts for an additional 26% to 29% of the variance, characterized by substantial primary loadings from Items 6 through 8: Upper-arm function (.82), Hand movements (.91), and Advanced hand activities (.94). Postural and transfer items load negligibly on this dimension.

Confirmatory Factor Analysis and Model Fit

Confirmatory factor analytic investigations have compared a unidimensional single-factor model against the theoretical two-factor model (Gross Postural/Mobility vs. Upper Extremity Function). The two-factor oblique model consistently demonstrates superior fit indices:

• Root Mean Square Error of Approximation (RMSEA): .052 to .064 (indicating good approximate model fit; < .08 threshold satisfied).
• Comparative Fit Index (CFI): .97 to .99 (exceeding the rigorous .95 psychometric standard).
• Tucker-Lewis Index (TLI): .96 to .98.
• Standardized Root Mean Square Residual (SRMR): .038 to .044.

While Items 1 to 8 can be summed to generate a meaningful composite motor recovery metric (justified by a strong inter-factor correlation of $r pprox .68$), the factor structure confirms that upper-limb recovery and gross locomotor balance represent semi-autonomous functional neuroanatomical modules following cortical and subcortical ischemic damage.

10. Instrument / Measurement Tool

• Instrument Name: Motor Assessment Scale (MAS) for Stroke.
• Authors: Janet H. Carr, Roberta B. Shepherd, Lena Nordholm, and Denise Lynne (1985).
• Test Type: Clinician-administered, standardized physical performance assessment.
• Target Population: Adult and elderly individuals recovering from cerebrovascular accident (ischemic or hemorrhagic stroke), brain injury, or acute neurological trauma affecting central motor pathways.
• Administration Time: Approximately 15 to 30 minutes, depending on patient disability severity.
• Equipment Required: Standard clinical examination plinth/bed, standard armless chair, stopwatch/timer, small table, circular plastic cup/tumbler, rubber ball (approx. 14 cm diameter), pencil, paper, cylindrical object, small beans or coins, shelf.
• Structure: 9 items total. Items 1 to 8 assess physical motor performance; Item 9 evaluates systemic resting and dynamic motor tonus.
• Response Format: 7-point hierarchical Guttman-like scale per motor area (0 to 6, where 0 = unable to perform criterion 1, and 6 = optimal performance of the most difficult criterion in that task area). Item 9 (General Tonus) is scored on a separate qualitative/categorical scale from 1 to 6 (or assessed continuously throughout).
• Scoring and Composite Rules: Items 1 through 8 are scored from 0 to 6 according to progressive hierarchical criteria. The scores of the 8 motor items are summed to yield a total motor score ranging from 0 to 48. Item 9 (General Tonus) assesses overall body tone throughout the test and is not included in the total motor score.
• Hierarchical Administration Guideline: Testing within each item begins at level 1 and proceeds sequentially to level 6. Up to three trials per item are permitted, recording the best achieved performance score. Patients should be tested in a well-lit, non-distracting environment and instructed to attempt each movement without external assistance unless explicitly designated by the manual.

11. Permissions & Fee and Test Year

The Motor Assessment Scale was initially published in 1985 in the journal Physical Therapy (American Physical Therapy Association). The instrument was placed into the public domain for clinical and academic research purposes to foster standardized neurological outcome assessments worldwide.

• Test Year of Origin: 1985.
• Accessibility and Fees: The MAS is free of charge for non-commercial clinical, educational, and scientific research use. No proprietary software, specialized paywalled score forms, or licensing royalties are required to administer the scale.
• Copyright and Attribution: The original publication is copyrighted by the American Physical Therapy Association (APTA). Researchers and health organizations reproducing the instrument within publications, medical electronic records, or academic curricula must provide formal attribution to the original 1985 publication: Carr, J. H., Shepherd, R. B., Nordholm, L., & Lynne, D. (1985). Investigation of a new motor assessment scale for stroke patients. Physical Therapy, 65(2), 175–180.

12. References

Below is a curated list of seminal peer-reviewed literature validating the psychometric properties, factor structure, and clinical utility of the Motor Assessment Scale in accordance with APA 7th edition formatting:

• Carr, J. H., Shepherd, R. B., Nordholm, L., & Lynne, D. (1985). Investigation of a new motor assessment scale for stroke patients. Physical Therapy, 65(2), 175–180. https://doi.org/10.1093/ptj/65.2.175
• Carr, J. H., & Shepherd, R. B. (1987). A motor relearning programme for stroke (2nd ed.). Heinemann Physiotherapy.
• Dean, C. M., & Mackey, F. H. (1992). Motor Assessment Scale scores as a measure of rehabilitation outcome following stroke. Australian Journal of Physiotherapy, 38(1), 31–35. https://doi.org/10.1016/S0004-9514(14)60548-2
• Loewen, S. C., & Anderson, B. A. (1988). Reliability of the Modified Motor Assessment Scale and the Barthel Index. Physical Therapy, 68(7), 1077–1081. https://doi.org/10.1093/ptj/68.7.1077
• Malouin, F., Pichard, L., Bonneau, C., Durand, A., & Corriveau, H. (1994). Evaluating motor recovery in stroke patients: A comparison of the Motor Assessment Scale and the Fugl-Meyer Assessment. Physical Therapy, 74(4), 284–294. https://doi.org/10.1093/ptj/74.4.284
• Poole, J. L., & Whitney, S. L. (1988). Motor Assessment Scale for stroke patients: Concurrent validity and interrater reliability. Archives of Physical Medicine and Rehabilitation, 69(3 Pt 1), 195–197.
• Poole, J. L., & Whitney, S. L. (2001). Assessments of motor function poststroke: A review. Physical & Occupational Therapy in Geriatrics, 19(2), 1–22. https://doi.org/10.1080/J148v19n02_01
• Sabari, J. S., Lim, A. L., Velozo, C. A., Lehman, L., Marks, P., & Capela, A. S. (2005). Assessing arm and hand function after stroke: A validity test of the hierarchical scoring system used in the Motor Assessment Scale. Archives of Physical Medicine and Rehabilitation, 86(8), 1609–1615. https://doi.org/10.1016/j.apmr.2005.03.013

13. Items of the Scale (Questionnaire)