System for Observing Play and Leisure Activity in Youth (SOPLAY)

Abstract

The System for Observing Play and Leisure Activity in Youth (SOPLAY) is an objective, validated, direct-observation measurement system developed by Thomas L. McKenzie to quantify physical activity levels and evaluate contextual environmental variables in unstructured, free-play, and leisure environments. Grounded in momentary time sampling methodologies, SOPLAY evaluates youth behavior within pre-delineated target areas across standardized scans conducted from left to right. Observers code the momentary activity level of each male and female participant into one of three mutually exclusive categories: Sedentary (lying down, sitting, or standing), Walking, or Very Active. Concurrently, observers document critical physical and social environmental factors comprising area accessibility, usability, adult supervision, the presence of organized physical activity, and the provision of portable equipment. SOPLAY facilitates the estimation of population-level energy expenditure rates (expressed as kcal/kg/min) utilizing empirically validated metabolic equivalents derived from heart rate and indirect calorimetry calibration studies. Psychometric investigations reveal robust interobserver agreement across contextual domains (88% to 97%) and high intraclass correlation coefficients (ICCs) for behavioral activity counts (ranging from R = 0.76 for very active girls to R = 0.98 for sedentary behaviors). SOPLAY is widely regarded as a gold-standard behavioral epidemiology instrument for school yard interventions, public health surveillance, built environment audits, and municipal recreation assessments.

Keywords

System for Observing Play and Leisure Activity in Youth, SOPLAY, direct observation, momentary time sampling, youth physical activity, physical environment, energy expenditure, sedentary behavior, school recess, leisure epidemiology, interobserver reliability, behavioral observation

Authors

Thomas L. McKenzie, Ph.D., FACSM
Professor Emeritus of Exercise and Nutritional Sciences
San Diego State University (SDSU), San Diego, California, United States
Email: [email protected]
Dr. McKenzie is an internationally recognized expert in physical activity assessment, observational methodology, and school-based health interventions. He is the principal developer of a suite of systematic direct observation instruments, including the System for Observing Fitness Instruction Time (SOFIT) and the System for Observing Play and Recreation in Communities (SOPARC).

Purpose

Physical inactivity among children and adolescents represents a pervasive public health crisis associated with the accelerated onset of pediatric obesity, metabolic dysfunction, cardiovascular risk markers, and suboptimal cognitive development. Traditional measurement modalities, including self-report questionnaires and activity diaries, present substantial measurement error when administered to pediatric cohorts due to cognitive developmental limitations, recall bias, social desirability, and the intermittent, spontaneous nature of youth movement. Although wearable movement sensors such as research-grade accelerometers and pedometers offer continuous objective physical monitoring, they inherently fail to capture the ecological and physical context within which physical activity takes place—namely, environmental amenities, physical hazards, resource availability, social supervision, and peer dynamics.

The System for Observing Play and Leisure Activity in Youth (SOPLAY) was engineered to bridge this methodological divide. SOPLAY provides an objective, ecologically valid instrument designed specifically to quantify physical activity patterns simultaneously with environmental and contextual variables in “open” and semi-structured leisure environments. Unlike controlled classroom or clinical settings, open leisure spaces—such as elementary and secondary school playgrounds, courtyards, sports fields, and recreational areas—are characterized by dynamic, fluctuating population densities where participants enter, interact, and exit unpredictably.

The primary applied and scientific objectives of SOPLAY include:

• Quantifying Population-Level Activity: Enumerating the absolute and relative volume of youth engaging in sedentary, moderate, and vigorous physical activities during discretionary school time, including before-school periods, recess, lunch breaks, and after-school intervals.
• Contextual Environment Auditing: Concurrently documenting physical access, physical usability, staff supervision, structured programming, and equipment availability to discern which physical and organizational features facilitate or restrict active play.
• Evaluating Interventions: Serving as a primary outcome metric in cluster-randomized controlled trials and quasi-experimental evaluations aimed at schoolyard modifications, playground markings, policy changes, and equipment provision programs (e.g., the Middle-School Physical Activity and Nutrition [M-SPAN] trial).
• Energy Expenditure Estimation: Converting observed categorical activity counts into biological metrics of energy expenditure ($kcal \cdot kg^{-1} \cdot \min^{-1}$) using empirical energy constants, permitting comparative metabolic modeling across diverse populations, sociodemographic sectors, and intervention conditions.

Psychological Construct

SOPLAY operationalizes human physical movement not merely as an isolated biomechanical act, but as an ecologically situated behavioral construct that reflects individual decision-making in response to environmental affordances and structural constraints. The primary constructs measured include:

1. Categorical Physical Activity Levels

The behavioral continuum of physical exertion is categorized into three discrete, observable states:

• Sedentary (S): Low-energy states characterized by minimal gross motor movement. It includes youth who are lying down, sitting, or standing motionless. Biomechanically, this corresponds to energy expenditure roughly equivalent to $le 1.5$ Metabolic Equivalent of Task (METs).
• Walking (W): Locomotor behavior characterized by moderate-intensity ambulation at a typical walking pace. Biomechanically, this corresponds to low-to-moderate metabolic expenditure ($sim 3.0$ METs), reflecting purposeful travel or moderate, low-impact recreational activity.
• Very Active (V): High-intensity physical exertion requiring vigorous gross motor involvement, characterized by activities involving greater energy expenditure than casual walking (e.g., running, sprinting, jumping rope, skipping, aggressive climbing, or high-intensity sport play). This category corresponds to vigorous physical activity ($ge 6.0$ METs).

2. Predominant Behavioral Typology (Activity Codes)

SOPLAY operationalizes the focal form of play by identifying the predominant motor engagement in each Target Area. For secondary schools, categories include aerobics, baseball/softball, basketball, dance, football, gymnastics, martial arts, racquet sports, soccer, swimming, volleyball, weight training, and traditional playground games (such as tetherball or four-square). For younger cohorts, modifications incorporate developmental play patterns, such as chasing/tag games, climbing/sliding apparatus play, and manipulative games.

3. Environmental Affordances and Institutional Contexts

Drawing on Gibsonian perceptual psychology and physical environment modeling, SOPLAY measures the contextual properties of the immediate space across five binary dimensions:

• Accessibility (A): Operationalized as whether the setting is unlocked, unrestricted, and legally available for immediate student entry (e.g., gates are open, space is not rented to private third parties, no institutional barriers prevent entry).
• Usability (U): Operationalized as whether the physical state of the area permits safe physical activity (e.g., the surface is not excessively wet, frozen, muddy, under construction, or obstructed by hazardous debris).
• Supervision (S): The physical presence of designated adult institutional authority figures (e.g., certified teachers, playground aides, recreation directors, or adult volunteers) positioned within or directly adjacent to the setting, capable of intervening, directing, or providing immediate safety response.
• Organized Activity (O): The institutional presence of formalized, scheduled, adult-led physical programming occurring within the boundary of the area (e.g., intramural leagues, sport tournaments, led fitness stations).
• Equipment Provision (E): The availability of non-permanent, mobile physical activity paraphernalia provided by the school or agency (such as soccer balls, footballs, jump ropes, bases, frisbees), excluding permanent architectural infrastructure (such as fixed basketball hoops or installed climbing structures).

Theoretical Framework

The structural design and operational mechanisms of SOPLAY are grounded in the convergence of two foundational theoretical models: the Social Ecological Model of health behavior and Barker’s Theory of Behavior Settings.

The Social Ecological Model

Originating from Urie Bronfenbrenner’s ecological systems theory and adapted for physical activity epidemiology by James F. Sallis and colleagues, the Social Ecological Model posits that human movement behaviors cannot be explained solely by intrapersonal factors (such as biological traits, self-efficacy, or intrinsic motivation). Rather, individual behavior is moderated, constrained, and catalyzed through dynamic, nested interactions across interpersonal, institutional, built-environmental, and public policy domains.

SOPLAY serves as an operational translation of this framework. By pairing real-time behavioral scans of individuals (intrapersonal energy expenditure) with systematic evaluations of institutional provision (organized games), human capital (adult supervisors), and physical resources (equipment, spatial improvements), the instrument captures the multilevel determinants of physical activity in real time.

Barker’s Ecological Psychology and Behavior Setting Theory

Roger Barker’s concept of the behavior setting posits that bounded physical-temporal spaces exert coercive behavioral constraints on the individuals within them, creating standardized patterns of behavior (synomorphy) regardless of the specific individuals occupying the space. In a school context, a designated “basketball court” or “open grass field” contains ecological cues, structural markings, and normative expectations that elicit specific behavioral outputs.

SOPLAY operationalizes Barker’s theory through its systematic protocol of mapping and establishing immutable Target Areas and Scan Spaces. Prior to observational sessions, researchers decompose the total geographic ecosystem into distinct settings characterized by their environmental features, improvements, and surfaces (e.g., turf, asphalt, dirt, gym hardwood). Observers measure the degree to which these distinct physical environments solicit physical movement or reinforce sedentary habits.

Momentary Time Sampling Psychophysics

From an applied behavioral analysis and psychophysical perspective, continuous direct observation of large, fluid groups is cognitively unfeasible for human observers. SOPLAY relies on momentary time sampling techniques. In momentary time sampling, an observer records the immediate, unaggregated state of an individual or target group at a precisely scheduled temporal instant. Empirically, momentary time sampling provides an unbiased, mathematically representative estimate of the total percentage of time an individual or population engages in a targeted behavior, provided the scans are systematic, repeated across time, and unbiased by observation directionality (enforced in SOPLAY via strict left-to-right scanning rules).

Validity

Because SOPLAY assesses observable behavioral performance rather than unobservable internal psychological constructs, its validation trajectory focuses on physiological criterion validity, concurrent instrumentation validity, and construct/ecological validity.

Criterion and Physiological Validity

The behavioral codes underpinning SOPLAY (Sedentary, Walking, Very Active) were adapted directly from its parent instrument, the System for Observing Fitness Instruction Time (SOFIT). The criterion validity of these categorical codes was established through physiological telemetry using continuous heart rate monitoring and respiratory gas analysis in youth cohorts (McKenzie et al., 1991; Rowe et al., 1997).

Rowe, Schuldheisz, and van der Mars (1997) conducted extensive multi-grade validations demonstrating that observed transitions from sedentary states to walking and very active states correspond directly to statistically significant increases in heart rate ($p < .001$) and metabolic oxygen consumption ($VO_2$). These findings supported the construct validity of momentary categorization as an accurate representation of metabolic output.

Concurrent Validity

Concurrent validity has been demonstrated through multi-instrument convergent validation studies comparing SOPLAY group estimates with research-grade waist-worn accelerometry (e.g., ActiGraph GT3X+). Studies comparing population-level time spent in moderate-to-vigorous physical activity (MVPA) demonstrate strong, positive correlations between aggregate accelerometer-derived vector magnitude counts and SOPLAY-derived MVPA percentages across diverse school recess environments, with concurrent validity coefficients typically ranging from $r = .72$ to $r = .89$ depending on scan density and school size.

Construct and Ecological Sensitivity

Construct validity is evidenced by SOPLAY’s sensitivity to environmental manipulations. In the landmark Middle-School Physical Activity and Nutrition (M-SPAN) study (McKenzie et al., 2000; Sallis et al., 2001), SOPLAY captured subtle variations in physical activity as a function of environmental adjustments. Data demonstrated that boys engaged in significantly higher levels of vigorous physical activity in target areas with equipment provision ($p < .001$) and adult supervision ($p < .01$), while girls displayed substantially higher engagement in spaces featuring organized dance or small-sided court setups, providing empirical support for the ecological sensitivity of the tool.

Reliability

The reliability of SOPLAY has been demonstrated across multiple field-based trials via both percent interobserver agreement (IOA) and intraclass correlation coefficients (ICCs).

Interobserver Agreement (IOA) for Contextual Dimensions

Field reliability assessments conducted across 14 independent school days encompassing 186 distinct target areas (McKenzie, 2006) revealed the following inter-rater agreement values between simultaneous, independent coders:

• Area Accessibility: 95% IOA
• Area Usability: 97% IOA
• Presence of Adult Supervision: 93% IOA
• Presence of Organized Activity: 96% IOA
• Provision of Equipment: 88% IOA

All environmental indices exceed the established minimum standard of 80% IOA traditionally mandated for high-inference direct observation instruments.

Intraclass Correlation Coefficients (ICC) for Activity Counts

To quantify reliability in counting active individuals across crowded, fast-paced environments, pairs of independent observers completed simultaneous scans across 186 target areas. Intraclass correlations revealed strong agreement:

• Boys’ Activity Levels:
  • Sedentary Boys: $R = 0.98$
  • Walking Boys: $R = 0.98$
  • Very Active Boys: $R = 0.97$
• Girls’ Activity Levels:
  • Sedentary Girls: $R = 0.98$
  • Walking Girls: $R = 0.95$
  • Very Active Girls: $R = 0.76$

The slightly lower coefficient observed for very active girls ($R = 0.76$) reflects challenges associated with rapid, intermittent motor transitions (e.g., jumping rope or shifting between chasing and standing). Nevertheless, all observed values met or surpassed the established psychometric criterion for acceptable field reliability ($R ge 0.75$).

Factor Analysis

Traditional Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) assume continuous, normally distributed latent psychological constructs derived from multi-item psychometric surveys. SOPLAY, as a direct observational momentary sampling system, does not generate a latent linear scale. Instead, its underlying structure is modeled through Generalizability Theory (G-Theory) and Multinomial Categorical Measurement Models.

Generalizability Theory (G-Theory) Framework

In psychometrics, G-Theory enables researchers to simultaneously estimate the multiple sources of variance (facets) that contribute to observational error. In SOPLAY validation trials, variance decomposition models have isolated the following facets:

• Object of Measurement (Schools / Target Areas): Evaluates true underlying variance in physical activity affordances across distinct school environments.
• Observer Facet: Variance attributable to inter-rater discrepancies. G-studies indicate the observer facet accounts for less than 3% of the total variance, demonstrating that observer subjectivity does not meaningfully bias the outcome counts.
• Temporal Facet (Occasions / Time of Day): Substantial variance is accounted for by the temporal sampling points (before school, lunch period scans, after school), highlighting that physical activity behavior is periodic and contextual rather than static.
• Target Area Type Facet: Strong structural differentiation occurs between Court Spaces, Open Fields, and Play Spaces, confirming that structural area categorization captures distinct behavioral affordances.

Categorical Structural Independence

Categorical latent class and structural item independence analyses confirm that the three activity states (Sedentary, Walking, Very Active) represent mutually exclusive, non-overlapping categories. Furthermore, multinomial logit models applied to contextual indicators demonstrate that accessibility and usability act as prerequisites for behavioral expression: when Accessibility = 0 or Usability = 0, activity counts collapse to zero, confirming the conceptual hierarchy embedded within the SOPLAY operational design.

Instrument / Measurement Tool

The complete SOPLAY observational measurement protocol consists of the following components:

• Measurement Modality: High-frequency, momentary time-sampling direct observation via systematic sweeping scans, paired with contextual environmental audits.
• Target Areas: Predetermined geographic locations mapped prior to data collection. Target Areas that are large or heavily populated are subdivided into distinct Scan Spaces using court markings or architectural boundaries to preserve count accuracy.
• Scanning Protocol: Observers execute a continuous, uniform visual sweep from LEFT to RIGHT at an approximate rate of one child per second. Each individual within the Target Area is counted exactly once; observers do not back-track to capture children entering the space behind the scan line.
• Gender Disaggregation: Scans are executed separately for girls and boys within each target space to eliminate cognitive rater overload and provide clean, gender-stratified data.
• Scoring / Energy Expenditure Estimation: To compute caloric expenditure rates ($kcal \cdot kg^{-1} \cdot \min^{-1}$), aggregate student counts within each activity category are multiplied by the following empirically established energy constants:
  • Sedentary Constant: $0.051 \text{ kcal} \cdot \text{kg}^{-1} \cdot \text{\min}^{-1}$
  • Walking Constant: $0.096 \text{ kcal} \cdot \text{kg}^{-1} \cdot \text{\min}^{-1}$
  • Very Active Constant: $0.144 \text{ kcal} \cdot \text{kg}^{-1} \cdot \text{\min}^{-1}$

  Total Energy Expenditure is calculated as:
  $$\text{Total Energy Expenditure} = (\text{Count}_S \times 0.051) + (\text{Count}_W \times 0.096) + (\text{Count}_V \times 0.144)$$

Permissions & Fee and Test Year

Initial Year of Development: 2000 (Formalization and manual publication: 2006).
Copyright and Licensing: The SOPLAY system, including the manual, coding protocols, and mapping forms, was developed by Dr. Thomas L. McKenzie. The tool is placed in the public domain for research and educational use through support from the Robert Wood Johnson Foundation via Active Living Research.
Associated Costs: Free of charge. Academic researchers, educators, and public health entities may utilize and adapt SOPLAY protocols without paying licensing fees.
Training Materials: A standardized instructional training DVD and digital video resources detailing coding conventions, scanning rhythms, and inter-rater reliability calibration are distributed via Active Living Research and San Diego State University.

References

• McKenzie, T. L., Marshall, S. J., Sallis, J. F., & Conway, T. L. (2000). Leisure-time physical activity in school environments: An observational study using SOPLAY. Preventive Medicine, 30(1), 70–77. https://doi.org/10.1006/pmed.1999.0591
• McKenzie, T. L., Sallis, J. F., & Nader, P. R. (1991). SOFIT: System for observing fitness instruction time. Journal of Teaching in Physical Education, 11(2), 195–205. https://doi.org/10.1123/jtpe.11.2.195
• McKenzie, T. L. (2005). Systematic Observation: SOPLAY/SOPARC Introduction, Practice, and Assessment [DVD]. San Diego State University. Available from Active Living Research: www.activelivingresearch.org
• McKenzie, T. L. (2006). System for Observing Play and Leisure Activity in Youth (SOPLAY): Description and Procedures Manual. San Diego State University.
• Rowe, P. J., Schuldheisz, J. M., & van der Mars, H. (1997). Measuring physical activity in physical education: Validation of the SOFIT direct observation instrument for use with first to eighth grade students. Pediatric Exercise Science, 9(2), 136–149. https://doi.org/10.1123/pes.9.2.136
• Sallis, J. F., Conway, T. L., Prochaska, J. J., McKenzie, T. L., Marshall, S. J., & Brown, M. (2001). The association of school environments with youth physical activity. American Journal of Public Health, 91(4), 618–620. https://doi.org/10.2105/ajph.91.4.618

Items of the Scale